Volume 29 Number 4, Winter 1986/87 ; is anging eiimaie ISSN 0029-8182 Oceanus The International Magazine of Marine Science and Policy Volume 29, Number 4, Winter 1986/87 ^^^^40G;?^^^ Paul R. Ryan, Editor James H. W. Hain, Assistant Editor Eleanore D. Scavotto, Editorial Assistant Lucia Susani, Fall Intern Editorial Advisory Board 1930 Henry Charnock, Professor of Physical Oceanography, University of Southampton, England Edward D. Goldberg, Professor of Chemistry, Scnpps Institution of Oceanography Gotthilf Hempel, Director of the Alfred Wegener Institute for Polar Research, West Germany Charles D. Hollister, Dean of Graduate Studies, Woods Hole Oceanographic Institution John Imbrie, Henry L. Doherty Professor of Oceanography, Brown University John A. Knauss, Provost for Marine Affairs, University of Rhode Island Arthur E. Maxwell, Director of the Institute for Geophysics, University of Texas Timothy R. Parsons, Professor, Institute of Oceanography, University of British Columbia, Canada Allan R. Robinson, Gordon McKay Professor of Geophysical Fluid Dynamics, Harvard University David A. Ross, Chairman. Department of Geology and Geophysics, and Sea Grant Coordinator, Woods Hole Oceanographic Institution Published by Woods tlole Oceanographic Institution Guy W. Nichols, Chairman, Board of Trustees James S. Coles, President of the Associates John H. Steele, President of the Corporation and Director of the Institution The views expressed in Oceanus are those of the authors and do not necessarily reflect those of the Woods Hole Oceanographic Institution. Permission to photocopy for internal or personal use or the Internal or personal use of specific clients is granted by Oceanus magazine to libraries and other users registered with the Copyright Clearance Center (CCC), provided that the base fee of $2,00 per copy of the article, plus .05 per page is paid directly to CCC, 21 Congress Street, Salem, MA 01970. Special requests should be addressed to Oceanus magazine. ISSN 0029-8182/83 $2.00 + .05 Editorial correspondence: Oceanus magazine. Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543. Telephone (617) 548-1400, ext. 2386. Subscription correspondence, U.S. and Canada: All orders should be addressed to Oceanus Subscriber Service Center, P.O. Box 6419, Syracuse, N.Y. 13217. Individual subscription rate: $22 a year; Libraries and institutions, $50. Current copy price, $5.00—25% discount on current copy orders for 5 or more; 40% discount to bookstores and newsstands. Please make checks payable to Woods Hole Oceanographic Institution. Subscribers outside the U.S. and Canada, please write: Oceanus, Cambridge University Press, the Edinburgh Building, Shaftesbury Rd., Cambridge CB2 2RU, England. Individual subscription rate £20 a year; Libraries and Institutions, £37. Single copy price, £9. Make checks payable to Cambridge University Press. When sending change of address, please include mailing label. Claims for missing numbers from the U.S. and Canada will be honored within 3 months of publication; overseas, 5 months. - u Give Gift of the Sea ^t..wop»^*. This Season come aboard yourself now! Oceanus Domestic Subscription Order Form: U.S. & Canada* Please make checks payable to Woods Hole Oceanographic Institution. The International Magazine of Marine Science and Policy Please enter my subscription to OCEANUS for Published by Woods Hole , j j i ^ ' , .. , Individual: Oceanographic Institution ^ D one year at $22.00 D two years at $39.00 D \hree years at $56.00 Library or Institution: D one year at $50.00 D payment enclosed. (we request prepayment) n bill me Please send MY Subscription to: Please send a GIFT Subscription to: Name (please print) Name (please print) Street address Street address City State Zip City State Zip Foreign subscribers please use form inserted at last Donor's Name, page. 12/86 Address- Former NASA Administrator and WHO! Associate Director for Research. 94 Book Reviews 100 Index COVER: The Earth rising over the moon. (Photo courtesy NASA). BACK COVER: Gulf Stream temperature composite of 35 satellite passes obtained during first week of April 1984. (Image courtesy of NASA) Copyright* 1986 by the Woods Hole Oceanographic Institution. Oceanus (ISSN 0029-8182) is published in March, June, September, and December by the Woods Hole Oceanographic Institution, 93 Water Street, Woods Hole, Massachusetts 02543. Second-class postage paid at Falmouth, Massachusetts; Windsor, Ontario; and additional mailing points. POSTMASTER: Send address changes to Oceanus Subscriber Service Center, P.O. Box 6419, Syracuse, N.Y. 13217. 1 HAS THE SUBSCRIPTION COUPON BEEN DETACHED? If someone else has made use of the coupon attached to this card, you can still subscribe. Just send a check — $22 for one year (four issues), $39 for two, $56 for three — to this address: Woods Hole Oceanographic Institution Woods Hole, Mass. 02543 Please make checks payable to Woods Hole Oceanographic Institution ^i-^°PUp, 1930 PLACE STAMP HERE Oceanus Woods Hole Oceanographic Institution Woods Hole, Mass. 02543 wooas noie, fv\assacnubeub uzdho. i eiepnunc- \ui n j-^u- <~t^Kj, <-^i. ^_.ou. Subscription correspondence, U.S. and Canada: All orders should be addressed to Oceanus Subscriber Service Center, P.O. Box 6419, Syracuse, N.Y. 13217. Individual subscription rate: $22 a year; Libraries and institutions, $50. Current copy price, $5.00—25% discount on current copy orders for 5 or more; 40% discount to bookstores and newsstands. Please make checks payable to Woods Hole Oceanographic Institution. Subscribers outside the U.S. and Canada, please write: Oceanus, Cambridge University Press, the Edinburgh Building, Shaftesbury Rd., Cambridge CB2 2RU, England. Individual subscription rate £20 a year; Libraries and Institutions, £37. Single copy price, £9. Make checks payable to Cambridge University Press. When sending change of address, please include mailing label. Claims for missing numbers from the U.S. and Canada will be honored within 3 months of publication; overseas, 5 months. (goraHcemti m^dimm mffM-h K Y^ -<■ . 4 N 1 Ti-J ^1 1 *■ ' » / / 1 ^=^if^^^^w= ::a(*f--?=t=£ i^^Ppfct ^-^: |Plonktonic glgoef' 2 Introduction: The Oceans, Climate, and Technology by Francis P. Bretherton The study of climate change challenges scientists to understand the complex interactions between the atmosphere, the ocean, and the Earth. 9 The Oceans, Carbon Dioxide, and Global Climate Change by Berrien Moore III, and Bert Bolin The processes that govern the largest sink for carbon dioxide — the oceans. 16 Global Ocean Flux fay lames /. McCarthy, Peter C. Brewer, and Gene Feldman The ocean's biogeochemical cycles are a key element in understanding climate. 27 The Oceans as a Source of Biogenic Gases by Meinrat O. Andreae The oceans play a central role in regulating the emission of biogenic gases. 36 Man's Great Geophysical Experiment: Can We Model the Consequences? by Kirk Bryan Models of the atmosphere /ocean provide clues to climate change. 43 Orbital Geometry, CO2, and Pleistocene Climate by Nicklas C. Pisias, and John Imbrie Variations in the Earth's orbit influenced past climate changes. 50 The Polar Ice Sheets: A Wild Card in the Deck? by Stanley S. Jacobs 1986 Antarctic iceberg calving events may not mean ice sheet melting. 55 Polar Ice Cores by lulie M. Palais Ice cores preserve climate information dating back thousands of years. 62 Photo Essay: Humans in Ice Age Europe— 10,000 to 35,000 Years Ago 64 Pollen in Marine Cores: Evidence of Past Climates by Linda E. Heusser Pollen records in marine sediments are a source of past global climate data. 71 Forests and Climate: Surprises in Store by George M. Woodwell Destruction of forests adds to the increase in atmospheric CO2. 76 Spaceborne Observations in Support of Earth Science by D. lames Baker, and W. Stanley Wilson Use of satellites has stimulated the need to study the Earth's processes- physical, biological, chemical, and geological — on a global scale. 86 Profile: Robert A. Frosch Unafraid to Take Risks by lames H. W. Hain Former NASA Administrator and WHOI Associate Director for Research. 94 Book Reviews 100 Index COVER: The Earth rising over the moon. (Photo courtesy NASA). BACK COVER: Cult Stream temperature composite of 35 satellite passes obtained during first week of April 1984. (Image courtesy of NASA) Copyright® 1986 by the Woods Hole Oceanographic Institution. Oceanus (ISSN 0029-8182) is published in March, )une, September, and December by the Woods Hole Oceanographic Institution, 93 Water Street, Woods Hole, Massachusetts 02543. Second-class postage paid at Falmouth, Massachusetts; Windsor, Ontario; and additional mailing points. POSTMASTER: Send address changes to Oceanus Subscriber Service Center, P.O. Box 6419, Syracuse, N.Y. 13217. 1 o, 'ur pres- ent knowledge of the in- terior of the Sun does not allow for ac- curate predictions of solar luminosity (assumed to be proportional to the total solar irradiance measured at the Earth over long time scales). We therefore depend on high-precision observations of the solar irradiance over time scales of climatological significance to understand the Sun's role in climate var- iability. A modern high-precision database on solar total irradiance was started in 1980. The first 5 years (1980 to 1985) of total solar irradiance observations by a monitor on board the Solar Maximum Mission spacecraft show a clearly defined downward trend of —0.019 percent per year. Changes in irradiance of as little as 0.5 percent per century can cause the complete range of climate variations that have occurred in the past, ranging from ice ages to global tropical conditions. — R. C. Wilson and others in Long-term Downward Trend in Total Solar Irradiance, Science, 28 November 1986. r\ ^ ■■> . it Introduction: The Oceans, Climate, and Technology by Francis P. Bretherton li uman activities are changing our global environment, in ways that we dimly perceive, and do not fully understand. Over thousands of years of human civilization, we have adapted to this world in which we live, to its climate and vegetation, and to the bounty of its oceans and soil. We have learned to expect the fluctuations of daily weather and the cycle of the seasons, and we use these expectations in growing our crops, designing our houses, crossing the oceans, and enjoying our leisure. Certainly, floods, droughts, earthquakes, volcanos, pestilence, and other natural disasters can interrupt the routine of daily life, but we still anticipate a return to normal, and hope for the familiar environment upon which we have built our lives. Yet, evidence is accumulating that with our modern civilization and technology we are no longer passive spectators to this natural drama, but instead are active participants. Through our energy consumption, agricultural practices, and use of natural resources, we are contributing to changes in the global environment with profound yet largely unknown consequences, over the span of a few human generations. The steady increase in the concentration of carbon dioxide in the atmosphere — recorded for the last 30 years — is attributed to the burning of coal, oil, and other fossil fuels on a worldwide basis. The atmospheric concentration is moderated primarily by Figure /. The greenhouse effect. Short-wave radiation from the Sun reaches the Earth's surface unhindered, but the outgoing long-wave radiation is partially trapped, or retained, by carbon dioxide and other gases in the atmosphere. ocean uptake and by storage of carbon in vegetation on land. If the increasing trend continues, however, it will lead to a doubling sometime before the middle of the next century. CO2 is a greenhouse gas, that is, it acts like a blanket to warm the Earth (Figure 1). Incoming rays from the sun reach the surface unhindered, but the outgoing longer wavelength radiation is absorbed in the atmosphere, changing the heat balance for the entire system. Our best computer models predict that the result of a doubling of atmospheric carbon dioxide would be a global warming of 2 to 3 degrees Celsius, with about twice that magnitude in high latitudes, and with substantial, though less well determined, changes in rainfall patterns everywhere. These effects may not seem large, but they are comparable to those that have already occurred in the 18,000 years since the last ice age, when the world was a very different place in which to live. Furthermore, the present change may be accelerated, and would take place over only a few decades to a century. The trend seems irreversible, and our understanding of its implications is sketchy at best. We anticipate, however, that climate and our natural environment will move outside the range of historical experience. Our children and grandchildren throughout the world will have to develop new definitions of what is normal, adjust their expectations of floods and droughts, and alter their supplies of food and water. There will be new realities. Some of the changes may be beneficial, some may be harmful, and others may be of no consequence. However, unless they are perceived and understood in a timely manner, almost all are likely to be painful. At least we must attempt to lay the basis of knowledge and understanding on which our successors can act. Burning fossil fuel is not the only cause for concern. Other trace greenhouse gases are also increasing in the atmosphere, notably methane, or marsh gas. One suspected cause is greater emission due to the enhanced productivity of rice paddies* associated with the green revolution. Another * Since plants in rice paddies spend a goodly part of the year underwater, their metabolism is anoxic, and a production of methane results. The precise quantity is unknown. Figure 2. The Earth system for modeling global climate change Figure 2. The Earth system for modeling global climate change. The Deep South experienced a severe drought in the summer of 1986. Here, Paul English at Montrose, Georgia, Inspects a stunted (normally 8-toot high) corn crop. (UPl/Bettmann Newsphotos) possibility is competition from urban air pollution in the chemistry of the hydroxyl radical, which provides the primary mechanism tor removing methane from the atmosphere.* Together with man- made fluorocarbons and nitrous oxide from soil bacteria,** the effects on global climate are expected to be comparable to those of carbon dioxide. Other accompaniments of increasing population, such as soil erosion, desertification, and deforestation, probably interact with climate change and global biological productivity in ways that we have barely begun to perceive. The Challenge These issues present scientists with a profound challenge. The fragmentary evidence presently available attests to the importance of the complex interactions and feedbacks between physical and chemical processes in the atmosphere, the ocean, the solid earth, and all living organisms. Addressing these issues requires obtaining an understanding of the entire Earth system, by describing how its * The hydroxyl radical ( — OH) is produced mostly through the photochemical decomposition of water. Urban air pol- lution may react chemically with the available hydroxyl rad- ical, leaving less for methane removal. ** It is known that the amount of nitrous oxide (N.O) is increasing in the atmosphere. Nitrous oxide is produced by bacteria. It is speculated that a general increase in the use of fertilizers increases the role of the bacteria in the soil that emit N2O. component parts and their interactions have evolved, how they function, and how they may be expected to continue to evolve. A particular need is to develop the capability to predict those changes (both natural and human-induced) that will occur in the next decade to century. To obtain such understanding will take a program of research over several decades, bringing to bear all the tools and experience that can be drawn from every discipline of Earth Science. Though focused on major global interactions and feedbacks, it must be solidly based upon quantitative understanding of the specific processes contributing to all aspects of climate change and its interaction with the natural environment, and of the history of such changes and interactions as revealed in geological and other records. Since experience shows that new insights will come from most unexpected directions, new research must supplement, but not replace, studies motivated by sheer curiosity and more traditional utilitarian concerns. The problems are fundamental and the complexities daunting, but to achieve our goal we must begin now. A Systems View Meeting this challenge requires the integration of knowledge from many different disciplines and information from many different types of observation into a coherent view of the entire Earth system. To help focus on what is involved, Figure 2 shows a Mount St. Helens spewing volcanic ash into the atmosphere in the spring of 1980. Eruptions such as these are thought to play a role in climatic changes. (UPl/Bettmann archive) concept for a computer model suitable for describing global changes on time scales of decades to centuries. Though the model as a whole does not now exist, the individual components have all been developed to various degrees, and researchers are beginning to connect them one by one within such a framework. Though present computing capability still imposes substantial limitations, such quantitative models provide a tool for summarizing understanding of processes and interconnections in terms that are communicable to individuals in other disciplines, and are invaluable for testing against observations/hypotheses about both the functioning of individual components and the entire system. Figure 2 is divided into two main sections, the Physical Climate System, and Biogeochemical Cycles, surrounded by ovals representing various inputs and outputs. Within these sections the smaller labelled boxes such as Atmospheric Physics and Dynamics, or Ocean Dynamics, denote subsystems which should be thought of as groups of computer subroutines. The arrows and associated variables, such as Wind Stress or Heat Flux, denote the flow of information between the different subsystems necessary to describe their interactions. Each subsystem is described by a set of state variables, such as temperature distribution, or the number of individuals of various species. Each subroutine within the subsystem is an algorithm* quantifying present understanding of some process in terms of changes in the state variables. The input variables necessary to complete the subsystem description define the interactions with other subsystems. For example, most models of ocean circulation require as input the atmospheric wind stress, and the net heat and fresh water fluxes at the surface, but compute as output the sea-surface temperature (SST). On the other hand, atmospheric models require the SST as input, but have the capacity to compute the wind stress and fluxes of heat and fresh water. Figure 2 was constructed by identifying distinct communities of modelers and asking individuals in each community what input they require to make their models run. Because formulations of algorithms within a subsystem are more varied, the process descriptors within each box, such as Sea Ice or Marginal Seas, are more schematic, and no attempt has been made to list their individual inputs and outputs. Besides showing the relationship of various components to the whole, this conceptual model is also useful as an organizing concept for a more general discussion and for an assessment of the present state of our knowledge of the system. Discussion of the Conceptual Model FHuman activities such as the release of carbon dioxide into the atmosphere or changing patterns of land use, are inputs treated in this model only in scenario mode. Also, to discuss usefully the many impacts of climate and changing natural vegetation on human affairs, it will be necessary to add studies that include social, economic, and political factors. Present confidence in our computer models is greatest in the area of Atmospheric Physics and Dynamics, and least in Marine Biogeochemistry and Terrestrial Ecosystems. Since we know that the processes described in these last two biological subsystems are fundamental to determining concentrations of carbon dioxide and other greenhouse gases and hence changes in the physical climate system, particular attention must be given to nurturing research in these areas to improve our quantitative understanding. Figure 2 also shows key inputs from studies of deep sea sediments and solar system dynamics of information about the distribution in the past of land and ice and about the well known small changes in incoming solar radiation first elucidated by Milutin Milankovitch, a Yugoslavian geophysicist working in the 1920s and 1930s. It also shows as output records of ocean surface temperature that can be inferred from the species composition of phytoplankton skeletons in deep sea sediments, of land vegetation from pollens, and of atmospheric carbon dioxide from cores drilled in the polar ice sheets. In principle, these time histories can provide tests for overall model performance under substantially different conditions from the present, and hence * A step-by-step procedure for solving a problem. Eddies on all spatial scales are characteristic ot the tluid earth. Here clouds in the atmosphere reveal an eddy circulation pattern sifvilar to what one finds in the ocean. their documentation is a prerequisite to establishing credibility tor predictive purposes. Indeed, recent studies of sediments and ice cores have revealed that over the last million years ice ages have apparently been closely correlated with the Milankovitch cycles,* strongly suggesting that their ultimate cause is external to the system. On the other hand, data from ice cores have shown that cold periods also were associated with major reductions in atmospheric carbon dioxide, in themselves sufficient to cause the fluctuations in climate that were observed. To explain this paradox it appears necessary to suppose that changes in the ocean circulation in high latitudes (where the subtle Milankovitch effects are largest) drastically altered the biogeochemical pump that determines the natural equilibrium concentration of carbon dioxide in the atmosphere, generating a strong positive feedback loop that greatly amplified the original change in the radiative balance. Still more striking, after the most recent ice age, this alteration seems to have occurred in only 200 years. These unexpected observations remind us that our present understanding of how the system functions is subject to considerable revision, and that continual interplay between attempts at quantitative prediction and * The theory which attributes regular cycles in sunshine and climate to periodic changes in the Earth's axis, orbit, and precession (season of closest approach to the Sun). independent measurements is at the heart of advances in scientific understanding. Long-Term Measurements Also vital to achieving the ability to predict is to document unequivocally the changes that are actually taking place now and over the next several decades. Particularly important are records of variables on the interfaces between the various subsystems, such as surface temperature, or concentrations of greenhouse gases, and of forcing functions such as the input of solar radiation. It is distressing that 70 million observations recorded during the last 100 years are proving inadequate to distinguish the rise in global sea-surface temperature that is assumed to have already occurred because of documented increases in atmospheric carbon dioxide. Biases due to changes in methods of taking temperatures have introduced ambiguities that are comparable to the expected signal. This example shows how difficult it is to assure adequate accuracy and sampling on an ongoing basis. It will take a major effort by caring scientists and operational organizations to institute and sustain the required measurements and the information system to support them. New Observational Systems New techniques also are becoming available that permit for the first time a global view of key parts of the system. Observations of ocean color from an Satellites in place now or proposed for the future permit a global view, and global-scale data collection. contributions to understanding chemical processes in the stratosphere. In situ chemical analyses also are becoming increasingly sensitive, greatly enhancing our ability to detect and measure trace species throughout the system. In the ocean, moored current meters are routinely deployed for a year or more to obtain long- term records at one location. Buoys and drifters moving with the ocean currents can now be tracked by satellite independent of expensive ships. Such improvements promise more cost-effective measurements and better coverage of large-scale phenomena, and open new opportunities to observing both long-term changes and detailed processes. We Must Begin Now The prospect that over the next 50 years or so human activities will induce a major change in global climate requires an integrated approach to the study of the Earth as a system. Within a broad intellectual framework, emphasis should be on the interactions between the atmosphere, oceans, solid earth, and biota, and particularly those processes that contribute materially to global feedbacks in climate and vegetation. Key elements in such an approach are computer models capable of describing these processes and interactions in quantitative terms, long-term measurements and data systems to document the changes that are occurring, and observing systems that enable a global view. Though the difficulties are daunting, the challenge is great. To meet this challenge we must begin now. Francis P. Bretherton is a Senior Scientist in the Oceanography Section at the National Center for Atmospheric Research (NCAR), Boulder, Colorado. imager aboard the Nimbus 7 satellite flown by the National Aeronautics and Space Administration (NASA) have provided estimates of chlorophyll concentrations in surface waters throughout the North Atlantic, and indicate unexpectedly large areas of high productivity north of the Gulf Stream (page 23). The brief 3-month flight of Seasat in 1979 also provided glimpses of changes in sea level associated with near surface ocean currents worldwide, and of the potential for measuring routinely the wind stress at the ocean surface. Follow-on instruments for sea level and wind stress are planned for the early 1990s. In the same time frame, the Upper Atmosphere Research Satellite is expected to make major References Committee on Earth Sciences of the Space Science Board, National Research Council. 1982. A Strategy for Earth Science from Space in the 1980's, Part 1: Solid Earth and Oceans. Washington, D.C.: National Academy Press Committee on Earth Sciences of the Space Science Board, National Research Council. 1985. A Strategy for Earth Science from Space in the 1980's Part II: Atmosphere and Interactions with the Solid Earth, Oceans and Biota. Washington, D.C.: National Academy Press Earth System Sciences Committee, NASA Advisory Council. 1986. Earth System Science: Overview. Washington, D.C.: National Aeronautics and Space Administration U.S. Committee for an International Geosphere-Biosphere Program, National Research Council. 1986. Global Change in the Geosphere-Biosphere: Initial Priorities for an IGBP. Washington, D.C.: National Academy Press MnuNfl Lon. HnNnii MONTHLY nVERflGE CflRBON DIOXIDE CONCENTROT ION MLO-85 355 I |lllll| |IIIM|IIIII| |lllll|lllll| |lim|lllll|lllll|MIII|lllll|MIII|lllll|l 350 345 340 335 330 325 320 315 310 305 l[IIMI|lllll| |IIIII|MIII|MIII|IIIII|IIIII| |'""| The Oceans, Carbon Dioxide, Global Climate Change iliiii.lMlllliiinlllllllllMllllllllllllll IimiiImiiiI IlllllllMllllllllllMlllllllll I I I I I I f I I I I I I I I I I I I I I I I I I I I I I I I U I I I I I I I I I I I M I I III I I I I I 11 58 60 62 64 66 68 70 72 74 76 78 80 82 84 by Berrien Moore III, and Bert Bolin ^ix elements — carbon, nitrogen, phosphorus, sulfur, hydrogen, and oxygen — are of special interest in the study of our planet. Because of life, each of the first four elements follows a close loop or cycle through increasing molecular energy states, as the elements are incorporated into living tissue, and then decreasing energy levels as the tissues decompose. These cycles are an expression of life; in a sense, they are also the metabolic system for the planet. Their various patterns are the consequence of a myriad of biological, chemical, and physical processes that operate across a wide spectrum of time scales. In the absence of significant disturbance, these processes define a natural cycle for each element with approximate balances in the sources and sinks that result in a quasi-steady state for the cycle, at least on time scales less than a millennium. However, human activity since the beginning of the Industrial Revolution has increased to such an extent that it must now be regarded as a significant disturbance to these critical biogeochemical cycles. The magnitude of this activity is extensive and the effects are approaching an important stage: certain indicators of the state of particular cycles, such as the levels of atmospheric carbon dioxide and methane for the carbon cycle, have moved well outside recent historical distributions (Figures 1, 2). Similarly, changes in the nitrogen and sulfur cycles are reflected by the onset of what we now call "acid rain." Indeed, it is difficult to identify a major river or estuary that has not been affected by the addition of phosphate from agricultural, urban, or industrial sources. We, humans, have been eminently successful in using our mastery of science and technology to increase the production of food, fiber, and other essential goods to meet the needs of expanding populations. We have done this by altering natural and traditional patterns of land and water use and particularly by using large quantities of fossil fuel energy. But now, we must begin to consider the Above, Figure 1. Concentration of atmospheric CO2 in parts per million (ppm) of dry air versus time in years, observed with a continuously recording non-dispersive infrared gas analyzer at Mauna Loa Observatory, Hawaii. The smooth curve represents a fit of the data to an annual cycle which increases linearly with time. The dots indicate monthly average concentrations. (Courtesy C. D. Keeling, Scripps Institution of Oceanography.) 1720- 1700- 1680- 1660 1640 1620 Q) — ^ 3 O c o C — o - — n *^ «D _ »- 0) c'^ o t: C (0 O CL o - 1560- 1540- 1520 1500 H Barrow , Alaska 7 N Samoa 14 S 1640- Cape Kumukahi 1620- Hawaii 22°N 1600- 1580- 1560- __^^ " *" 1540- 6 1982 Time (years and months) Figure 2. Trends in methane (CH4J at four globally distributed locations. general scope of our industrial activities, the extent of which is shown not only by the Dow Jones industrial average, but also, for example, by the concentration of carbon dioxide in the atmosphere. The picture (Figure 1 ) of the concentration of CO2 in the atmosphere gives a remarkable portrait of the "natural" or, if you will, the ecological world overlain by the "industrial" or economic world. First, we see an oscillation, which is principally a reflection of the Northern FHemisphere's seasonal pattern of biological activity. Carbon dioxide decreases in the spring time as plants begin to "wake up" and take in carbon dioxide through photosynthesis which exceeds the return of carbon dioxide as a result of decomposition of organic matter in the soil. This continues through the summer, but with fall there is a reduction in photosynthetic activity, and carbon dioxide cycles upward. This continues further as plants drop their leaves, and the plant material begins to decompose and release additional carbon dioxide. Consequently, the atmospheric carbon dioxide peaks, only to decline again the following spring. Our "disruption" of this natural oscillating cycle of carbon by burning fossil fuels, harvesting forests, and converting land to agriculture, is reflected in the trend of increasing carbon dioxide concentrations that is superimposed on the seasonal pattern. The upward trend is significant: if it continues, the atmospheric concentration might exceed 600 parts per million by volume (ppmv) by the end of the next century — more than two times the pre-industrial level. The increase in carbon dioxide is interesting from a purely biogeochemical standpoint since it gives us an overall view of how the Earth works, but, since carbon dioxide is one of the "greenhouse gases," it is an important trend for another reason. The characteristic of carbon dioxide and water vapor — transparent to solar energy and yet absorbing some of the infrared radiation (heat) emitted by the Earth (emitted mainly as a result of solar radiation) — is one of the key phenomena that makes life possible on this planet (refer also to page 37). On the other hand, this can be overdone. It is primarily the high concentration of carbon dioxide in the atmosphere of Venus that causes that planet to be so terribly hot. We therefore need to pay attention to the increasing concentration of carbon dioxide in our planet's atmosphere. Generally, it is believed that an increase in the amount of carbon dioxide will result in an even greater atmospheric retention of radiant heat and a higher equilibrium temperature for the Earth as a whole. If atmospheric concentrations of carbon dioxide were to rise to 600 ppmv, then the increase in temperature is expected to be important: 1 .5 to 4 degrees Celsius for the entire Earth, with an 10 Deforestation 1-2 5-6 Fossil Fuel Atmosphere 740 j^'^>.. 50 Biota 550 u 60 70 80 22 35 ,,60' Soil and Detritus 1200 CARBON Reservoirs in 10 grams Fluxes in lO^^grams per years 10 grams = 1 billion metric tons Sedimentation 0.5 figure J. The global carbon cycle. Current estimates of the major reser\'oirs (in /O''' grams of carbon) and fluxes (in /O" grams of carbon per year) involved in the global carbon cycle. increase of approximately 5 to 10 degrees at the polar regions. More importantly, rainfall patterns might be affected dramatically, and this could alter the global distribution of agriculture. The Carbon Cycle — Unresolved Imbalances At present, our ability to interpret the carbon cycle (Figure 3) and, thus, predict future carbon dioxide concentrations in the atmosphere is confounded by unresolved imbalances in the carbon budget. Simply stated, the annual budget (Table 1) does not balance, and as yet unknown sinks, on land or in the oceans, must play a role. The imbalance may be diminished in several ways: by reductions in the estimate of the rate of deforestation or increases in the regrowth; if the oceanic uptake is uriderestimated; or, if there are significant shorter term natural variations in the concentration of carbon dioxide in the atmosphere that override the imbalances. The question of which c ombination of these possibilities is more likely is (:)oth important and fascinating. The present average concentration of carbon dioxide in the atmosphere is about 346 parts per million, which is equivalent to about 740 billion metric tons, or BMTs for short. This is up from less than 600 BMTs, which, based on ice core records, we believe was present in the atmosphere in the middle of the last century. Estimates of the amount of carbon in living organic matter on land vary between 450 and 600 BMTs. So there is more carbon in the atmosphere than there is in all the world's forest — a rather unbelievable fact. There are, of course, other "pools" of terrestrial carbon. For instance, there is two to three times more carbon in the surface soils of the world than in vegetation. The amount of carbon released globally from vegetation and soils as a result of deforestation remains uncertain, though much progress has been made in recent years. The difficulties in calculating Table 1. Annual Carbon Budget. Input of Carbon as Carbon Dioxide into Atmosphere Fossil Fuel: 5 Billion Metric Tons Carbon per year Deforestation Minus Regrowth: Total Input: 1 Billion Metric Tons Carbon per year 6 Billion Metric Tons Carbon per year Uptake of Carbon Dioxide Atmospheric Increase: Oceanic Uptake: Fertilization Effects: Total Uptake: 2.5 Billion Metric Tons Carbon per year 2.5 Billion Metric Tons Carbon per year ? 5 Billion Metric Tons Carbon per year + 11 If the world's temperature increases, rainfall patterns might be affected dramatically, thereby altenng the global distribution of agriculture. (The Bettmann archive) this release centers on two general but important factors: 1) the rate of land-use change, and 2) the response of biota to disturbances. It is surprising and frustrating that, even today, estimates of the current rate of conversion of closed canopy tropical forests to agricultural land vary from 70,000 to 100,000 square kilometers per year; the combined area of Massachusetts, New Hampshire, and Vermont is approximately 70,000 square kilometers. The analysis of the biotic response to such disturbances has proven even more difficult to determine accurately (see article page 71). Calculations of the net carbon loss from the global biotic inventory, which attempt to take into account the uncertainties concerning disturbance rates and biotic response, indicate that from 1860 to 1980 the total loss was between 100 and 200 BMTs, which is roughly the amount of carbon dioxide that the combustion of fossil fuels has contributed. So over the last 120 years or so, as much CO2 has come from deforestation as from fossil fuel combustion, though the present rates show that fossil fuel now is about three to five times more important (Figure 3). We should mention that this historical estimate of about 150 BMTs of carbon from deforestation is consistent with other calculations that use ratios of carbon isotopes, as found in ice cores and in tree rings, to infer how much CO2 has come from deforestation; however, the temporal pattern from these two types of calculations do differ somewhat. The oceans are by far the largest active reservoir of carbon. Recent estimates of the total amount of dissolved inorganic carbon establish an amount of about 38,000 BMTs. Only a small fraction is carbon dioxide (0.5 percent); the biocarbonate ion makes up to 90 percent and the carbonate ion, at slightly less than 10 percent, are the major forms of dissolved inorganic carbon. There is far less dissolved organic carbon, about 1,000 BMTs, and even less particulate organic carbon. Although the oceans are the largest active reservoir of carbon and cover almost 70 percent of the globe, the total marine biomass contains only about 3 BMTs of carbon, or just about 0.5 percent of the carbon stored in terrestrial vegetation. On the other hand, total primary production of marine organisms is 30 to 40 BMTs per year, corresponding to 30 to 40 percent of the total primary production of terrestrial vegetation (Figure 3). However, only a relatively small portion of this production results in particulate organic carbon, which sinks and decomposes in deeper layers or is incorporated into sediments. The net flux of atmospheric carbon dioxide into the oceans remains uncertain, although most current models suggest that between 40 and 50 percent of the net emissions to the atmosphere is taken up by the oceans. If these calculations reflect actual oceanic uptake, then, as the imbalance implies, this result is inconsistent with the magnitude of the fossil fuel emissions and biospheric source for CO2, the latter estimated either directly from land clearing patterns or indirectly from tree ring and ice core records of carbon isotopes (Table 1). To better appreciate the ocean's role in all of this, let us look a bit deeper into the particular biological, chemical, and physical processes that govern the largest sink for CO2: the ocean. The Role of the Oceans The rate of carbon dioxide uptake by the oceans is controlled by seawater temperature, surface chemistry, and biology, and by the various patterns of mixing and circulation which determine the amount of carbon transported from surface waters to the deep ocean. The actual exchange of CO2 between the sea surface and the atmosphere is by diffusion at the air/sea interface and hence is governed by the CO2 partial pressure difference between the ocean and atmosphere, the sea surface wind velocity, and the state of the ocean surface. The essential uncertainties are the distribution of the partial pressure of CO2 in the surface of the ocean, and the factors controlling this distribution. From the perspective of the global carbon cycle one could consider the partial pressure of CO2 in seawater as a linear function, depending on alkalinity and on the concentration of CO2 in seawater. Therefore, we primarily need to keep track of the CO2 and alkalinity in the sea surface. This is far easier said than done. First, CO2 dissociates in seawater, breaking into biocarbonate 12 and carbonate ions. Further complicating this chemical phenomenon are the dynamics of biological and physical processes. Primary production consumes COj; respiration and decay processes produce CO.. Each process affects the chemical equilibrium. Similarly, carbonate formation (shells) and dissolution alters alkalinity, which, as mentioned, also affects the partial pressure of CO, in seawater. The physical processes of circulation and mixing continually adjust the total inorganic carbon concentration throughout the ocean, and thereby continually change the COj partial pressure distribution in the surface waters. It is really quite a myriad of activity, and it is remarkable that oceanographers have pinned down the role of the ocean in the carbon cycle as well as they have! Five features about this role though do stand out: 1 ) the consumption of COj in primary production in biologically active surface waters; 2) the enrichment of the deep water in CO2 as the result of the decomposition and dissolution of detritial matter that originates from biological processes in the surface waters; 3) the sinking of water in polar regions, particularly in the North Atlantic, taking CO2 with it followed by a general bottom water flow toward the equator; 4) the upwelling of some of this water in equatorial regions with a corresponding outgassing of CO2 to the atmosphere and a general poleward flow of surface waters; and 5) accompanying the meridional circulation (oceanic overturning) are the general turbulent mixing processes; whereby, the carbon- rich water at intermediate depths are continuously being exchanged (mixed) with water of less carbon content in the surface layers. The first two features taken together are often referred to as the "biological pump"; the biology pumps carbon to the bottom. The most obvious pumping is the incorporation, in tissue or as carbonate in shells, into living organisms of COj that is dissolved in surface waters, lowering the partial pressure of COj, followed by the "shipping" of some of this CO2 to the bottom "packed" in the remains of dead marine organisms (Figure 4). As a consequence of the "biological pump," the concentration of dissolved inorganic carbon is not uniform with depth: the concentration in surface waters is 10 to 15 percent less than deeper waters. There is a corresponding depletion of phosphorus (and nitrogen) in surface water, even in areas of intense upwelling, as the result of biological uptake and the loss of the detritial material, which also contains phosphorus (and nitrogen) as well as carbon (Figure 5a and 5d). The fate of the carbon that falls from the surface waters depends, in part, upon its characteristics. If it is organic material, then it is oxidized at intermediate depths, which results in an oxygen minimum (Figure 5c) and a carbon and phosphorus maximum (Figure 5a and 5d). If the material is carbonate, it dissolves, raising both alkalinity (Figure 5b) and the concentration of carbon, primarily at great depths where the high pressure increases the solubility of calcium carbonate. Thus, where active, the biological pump Figure 4. A coccolHh, Coccolilhus huxleyi, one of the "gears" in the biological pump. (Transmission electron micrograph by Dr. Susumo Honjo. WHOI. X 12,000) lowers the partial pressure of CO2 in surface waters and enhances the partial pressure in deep water not in contact with the atmosphere. It is as if the "biological pump" moves the partial pressure around in a way that allows carbon dioxide to work its way into the ocean. In areas of low production, on the other hand, the partial pressure is often greater than that of the atmosphere, and CO2 is released from the sea surface. In the natural preindustrial steady state, however, the overall pattern was in balance; the net exchange was zero. Further, each change in alkalinity, phosphorus, total inorganic carbon, or oxygen is mediated by biological processes as well as by the motion of the water masses. By quantifying the biological processes and by exploiting oceanic chemical profiles (for example, data from the Geochemical Ocean Section Study and the Transient Tracers in the Ocean Program) marine geochemists have been able to determine the pattern of biological production and respiration, the strength of the biological pump, and, to a degree, the rates of circulation and mixing. The role of the oceans in the carbon cycle is much dependent on its rate of overturning (the meridional circulation) and its mixing. In polar regions, ice formation leaves much of the salt "behind" still in solution. The result is an increase in salinity in these already cold waters and hence an increase in density. As a consequence, these cold surface waters sink, and as such, they also have the potential to form, in effect, a pipeline or conveyor belt for transferring atmospheric CO2 to the large reservoirs of abyssal waters which have long residence times. This downward convection of surface waters in polar regions during "bottom water formation" creates a sink for carbon dioxide in high latitudes, but the balancing upwelling of carbon-rich waters in low latitudes creates a source. In other words, what goes down (cold polar water with CO2), must come 13 Distribution of dissolved inorganic carbon Dissolved inorganic carbon m Mol per kilo 2,00 2,10 2 20 1 ' 1 1 ' 1 1 Atlantic Ocean 200 2.10 2 20 2.30 ':'i Pacific Ocean India Ocea Ocean alkalinity Alkalinity m eq per kilogram 2.30 235 240 — "r ' ^ E '°°° Atlantic Ocean 2 30 2 35 Indian Pacific Ocean Ocean oxygen Oxygen m Mol per kilogram 015 20 0 25 0 30 Atlantic Ocea 0 5 0 20 0 25 C 30 dian Ocean Pacific Ocea Ocean phosphorus Phosphorus p Mol per kilogr 1 00 2 00 1 1 ' 1 , ,'■ Atlantic Ocean Indian Pacific Ocean ocean Figure 5. a) The observed distribution of dissolved inorganic carbon (mmol per kilogram) as a function of depth in the Atlantic, Pacific, and Indian Oceans, according to GEOSECS data, b) The observed distribution of alkalinity (meq per kilogram) as a function of depth, c) The observed distribution of oxygen (mmol per kilogram) as a function of depth, d) The observed distribution of phosphorus (mmol per kilogram) as a function of depth. 14 up (warm equatorial water with excess CO2). In addition to the bottom water formation in polar regions, there is water exchange between surface waters and intermediate waters resulting from vertical exchanges — a form of turbulent mixing or diffusion, in association with the surface ocean currents, like the Gulf Stream. These different exchange processes, bottom water formation and turbulent mixing, maintained by water motions renew the abyssal part of the oceans in the matter of a few hundred years in the Atlantic Ocean, while the age of Pacific deep water is up to about 1,500 years. Intuitively we realize that this rather slow rate of ocean turn-over limits the oceans as a sink for carbon dioxide. The Oceanic Sink For "Excess" CO2 There are two additional factors we need to consider in seeking to understand the oceans as a sink for the additional carbon dioxide that has been introduced into the atmosphere by humans. 1 . If atmospheric CO2 concentrations are enhanced by a given percentage, then only a rather small fraction need be transferred into the ocean in order to restore an equilibrium between the atmospheric CO2 pressure and the partial pressure of COj in the sea surface. It is as if COj responds with "inflated" value in the oceans with respect to partial pressure. This inflation factor, which we shall denote by R, is called the Revelle factor. As a matter of fact, the total amount of carbon need only be enhanced by a factor 1/R relative to the change of atmospheric COi pressure to achieve such a new balance, where R is between 9 and 14 depending upon ocean water temperature (and today's pattern of alkalinity). In other words, if 10 BMT of carbon dioxide were injected into the atmosphere, only approximately 1 BMT of CO2 would need enter the mixed layer of the ocean (which contains about the same amount of carbon as does the atmosphere) in order to re-establish the partial pressure equilibrium. This fact obviously markedly limits the ability of the ocean to take up the excess COj. 2. On the other hand, the interplay between the biological pump and the vertical transfer of carbon in the sea due to water actions does open the door just a bit and allows a larger net transfer of excess carbon dioxide. What allows this? What has changed in the ocean as the atmospheric COj has increased? First, the rate of primary production and detritus formation in the surface layers, and thus the rate with which the biological pump is operating, has probably not changed. What has changed is the gradient between the atmosphere and the oceans. This is a tricky area and we need to juggle three thoughts: a) The concentration of COj in the atmosphere has been increasing; certainly the concentration is greater today than, for example, 200 years ago. b) The waters upwelling today tend to reflect earlier "steady-state" conditions or at least atmospheric concentrations that are less than the present ones. In other words, the upward flow of carbon in these waters probably balanced the gross rate of CO2 uptake by a former ocean — in the preindustrial period; whereas, c) the gross flux of CO2 into the contemporary surface waters is greater than the corresponding flux for the preindustrial period simply because atmospheric concentrations, and hence atmospheric CO2 partial pressure, is greater today. In sum, there has been a decrease in the vertical difference of total carbon in seawater. This pattern of a rather slow oceanic turnover and mixing would appear to define a rather ponderous role for the oceans in the global carbon cycle. One might say that the sea appears to be only slowly responding to the marked perturbations of the atmospheric "pool" that humans are causing. From one point of view, this is true. It appears that only about half of the net input of CO2 to the atmosphere due to fossil fuel combustion and changing land use has so far been transferred into the oceans despite the fact that the size of the oceanic reservoir is more than 50 times that of the atmospheric one. But does this imply that we can neglect understanding further the ocean's role in the global carbon cycle? Quite the opposite. The implication is that even if we omit or misinterpret even rather minor processes in the ocean, this may have a significant influence on our view on what the likely future partitioning of excess CO2 will be. The ocean is somewhat akin to a national economy. It might be rather slow to respond, but misunderstanding what appear to be minor parameters can have costly effects. Important Questions Understanding these patterns of bottom water formation and the biological pump appear to be at the heart of understanding the global carbon cycle not only today, but even more importantly, in the future. If climate changes, how then will that affect the rate of bottom water formation or the biological pump? Will the ocean become more efficient or less efficient in storing CO2? These are important questions for Earth Science and vital issues for society. Berrien Moore III ;s Professor of Systems Research), Director of the Complex Systems Research Center, and a member of the Institute for the Study of Earth, Oceans, and Space at the University of New Hampshire. He is currently a visiting scientist at the Lahoratorie de Physique et Chemie Marines in Paris, France. Bert Bolin is Professor of Meterology and Dean of the Faculty of Mathematics and Natural Sciences at the University of Stockholm. He has been engaged in research on the carbon cycle for about 30 years. Since 1 983, he has been a member of the Scientific Advisory Group to the Swedish government, and, since early 1986, scientific advisor in the Prime Minister's Office. Selected Readings Bolln, B. 1970. The carbon cycle. Scientific American 223: 124-132. Bolin, B., E. T. Degens, S. Kempe, and P. Ketner, eds. 1979. The Global Carbon Cycle. 491 pp. New York: John Wiley and Sons. Carbon Dioxide Assessment Committee, National Research Council. 1983. Changing Climate. Washington, D.C.: National Academy Press. Trabalka, J.R., ed. 1985. Atmospheric Carbon Dioxide and the Global Carbon Cycle. 315 pp. Springfield, Virginia: United States Department ot Energy, Department of Commerce. 15 I he trend of increasing concentrations of carbon dioxide in the atmosphere, which is now firmly documented at several monitoring stations around the globe, builds on the data base initiated nearly 30 years ago at Mauna Loa, Hawaii, by Professor C. David Keeling of the Scripps Institution of Oceanography. Today, with help from the National Oceanic and Atmospheric Administration (NOAA) (see Figure 1, page 9, previous article), these data constitute one of the most important bodies of evidence linking the actions of humankind to global scale change in biogeochemical cycles. At present only a few of the effects of this change are known; many are predicted. Climate models predict that the increasing concentrations of "greenhouse" gases in the Earth's atmosphere will result in a global warming by as much as 2 to 3 degrees Celsius in the next half century. There has been evidence, however, to indicate atmospheric warming since early in the 20th century, and recent analyses of long time series for ocean surface temperature indicate a trend of ocean warming during the last 120 years (Jones and others, 1986, Nature 322:430-34). Change in responses to local shifts in weather and climate will probably influence agricultural productivity for some regions, yet the extent of this is, at present, impossible to predict. It is apparent that the largest changes in air temperature will be at high latitudes and the associated melting of polar ice predicted with climate models could raise sea level by about 1 meter. It should be noted, however, that melting of sea ice, such as that covering the Arctic Ocean, will have no direct effect on sea level, any more than melting ice will overflow a cocktail glass. It is the melting of ice accumulated on land, such as in Antarctica or Greenland that is at issue. Models used to predict such effects are becoming more refined through an improved understanding of the distribution of heat and momentum in the world's oceans. 16 >' V? The central Arctic Ocean is covered in summer by 8 million square kilometers (3 million square miles) of ice. But melting ot this sea ice has no direct affect on sea level any more than melting ice cubes effect the level of liquid in a glass of scotch and soda. SAR image of Arctic pack- ice motion here obtained by NASA's Seasat satellite. Marine Biogeochemical Exchange Processes During the last few years, the role of marine biogeochemical processes in the long term natural cycle of change in atmospheric CO2 content and its relationship to the periodic ice age cycles in climate has become more obvious. During the last ice age, about 18,000 years ago, the atmospheric COi content was strongly lowered, with a rapid rebound at the end of the glacial cycle. We know that the ocean must have played a dominant role in forcing the atmosphere, but we do not understand exactly how. However, the manner in which this coupling between biogeochemical cycles and climate might be affected by anthropogenic alteration might provide an important research focus in oceanography. It is no longer thought to be a simple coincidence that planet Earth, the "blue" planet, sometimes affectionately referred to as the "water- cooled" planet, has very different atmospheric composition from those of her sister planets. The great abundance of oxygen in the atmosphere of Earth (21 percent), which was a prerequisite to the evolution of animal life, and likewise the extremely low concentration of carbon dioxide (0.03 percent), are consequences of the photosynthetic activity by plants. In contrast, the atmospheres of Venus and Mars are nearly devoid of oxygen and contain 90 and 50 percent carbon dioxide, respectively. Water, even water vapor, is very scarce in the atmospheres of Venus (0.01 percent), and Mars (0.1 percent). A tempting scenario for the evolution of Earth's current atmosphere attributes not only the high levels of oxygen, but also the low levels of carbon dioxide to the great success of plants on our planet. It is obvious that it is not enough simply for plants to flourish. For each molecule of CO2 removed from the air by growing plants one is put back on decay of the tissue. We only see a net draw down if we squirrel away some of the fixed carbon by burial, and thus prevent the backflow to the atmosphere. Nature has done this for millions of years. Modern man is now reversing the process in a century or two. Plants on land and in the sea function similarly in their interaction with the atmosphere via the processes of photosynthesis and respiration. Animals and bacteria consume oxygen, and release carbon dioxide via the oxidation of assimilated organic material, originally synthesized by plants. These processes in the terrestrial system involve direct exchange between the organism and the atmosphere. In the marine system, and similarly in fresh water, the biologically mediated exchange is between the organisms and reservoirs of oxygen and carbon dioxide dissolved in water. Exchange of these gases between the atmosphere and the hydrosphere is controlled by physical and chemical processes. Since photosynthetic activity in oceanic waters can only occur at depths shallower than about 100 meters, while respiration occurs all the way to the seafloor, ocean mixing is key among the physical processes responsible for the air/sea exchanges of biologically active gases. It is the mixing of deep water to the ocean surface that permits carbon dioxide 17 The Mauna Loa Observatory on the Island of Hawaii. Key data obtained over the years at this station by Professor C. David Keeling of the Scripps Institution of Oceanography on the rising level of carbon dioxide in the atmosphere have served as benchmark observations in the field of climate studies. (Photo courtesy of David Moss, Scripps Institution of Oceanography) produced by respiration to escape and the oxygen depleted by this same process to be replenished. Within the terrestrial realm, the accumulation of carbon and depletion of oxygen occurs somewhat analogously in soils, but the long residence time of deep ocean waters and sediments results in much greater significance for the vast marine biogenic reservoirs in controlling the global carbon budget. Each species of plant has its own optimum set of conditions for growth and storage of carbon. The reverse process of decay is also regulated, with the time constants for the deep ocean being especially long. Thus, the accumulation of oxygen in the Earth's atmosphere partially reflects the different time constants for the processes of photosynthesis and oxidation of organic matter in terrestrial and marine ecosystems. Differences in Land, Sea Carbon Cycles Some of these key differences between the marine and terrestrial components of the carbon cycle are evident in Table 1. Whereas the biomass of terrestrial and marine organisms, which is predominantly plant material, differ by two orders of magnitude, the rates of photosynthetic activity or primary productivity for the two domains are rather similar. The values selected for these two rates are currently the subject of great debate, and many experts on this subject will argue that we know neither value to within a factor of two uncertainty. Explanation for the fact that the two rates can be similar while the quantities of biomass are so different lies in the contrast between the allocation of carbon in tissues of terrestrial and marine plants. Terrestrial plants contain large quantities of cellulose as structural material to support leaves, stems, and roots. In marked contrast, the dominant marine plants, unicellular phytoplankton, allocate only minor quantities of carbon to cellulose-like structural components, investing the bulk of their carbon in proteins, simple sugars, and lipids. This relatively small investment in organic structural materials allows for fast specific growth rates, with doubling times as short as one or two days for many of the marine phytoplankton. Many animals in the sea make shells of calcium carbonate, including those constructing vast coral reefs. However, only one group of phytoplankton, the coccolithophorids, use carbon Table 1. Major carbon reservoirs and fluxes. (Units in gigatons of carbon, where 1 Gt = 1 billion metric tons) The atmosphere of Venus, above, is nearly devoid of oxygen, but is made up mainly of carbon dioxide (90 percent). (Photo courtesy of NASA) RESERVOIRS GtC Atmosphere 700 Oceans: Total inorganic Particulate organic Land biota 35,000 3 600 Soil humus Marine sediments (Organic) (Calcium carbonate) Fossil fuels 3,000 10,000,000 50,000,000 5,000 FLUXES CtCyr-' Atmosphere — marine biota Atmosphere — land biota 45 70 Deposition in oceans Fossil fuel combustion 1-10 5 18 Plants on land and in the sea function similarly in their interaction with the atmosphere via the proctsscs ui phuLusynthesis and respiration. 19 dioxide dissolved in seawater to form calcium carbonate plates, known as coccoliths, that armor the cells' exterior (see page 13). While the biological processes contributing directly or indirectly to the flux of carbon dioxide from the atmosphere to the biosphere give rise to a roughly equal partitioning of the organic product between the marine and terrestrial realms, its fate, once organically fixed, differs greatly between the two. The greatest contrast between analogous components of these systems is in the reservoirs of remains for dead plants and animals, which consist mostly of remnants of structural materials. Whereas the quantity of carbon stored in soils is large compared to the standing stock of plants on land, the ratio of masses for the analogous reservoirs in the ocean is, by comparison, absolutely enormous. More than 99 percent of the carbon contained in the reservoirs influenced by biological processes now resides in marine sediments (Table 1, page 18). Although the calcium carbonate component of deep sea sediments is larger than the organic carbon component by a factor of five, even the ratio between the smaller of these two and the standing stock of ocean biota is immense relative to the analogous ratio for the terrestrial system. The rate at which carbon is now entering the deep ocean sediment reservoir, either as organisms or their calcium carbonate skeletons, is not known with great precision. Interestingly, the effect of adding COj from our atmosphere to the ocean is to make the naturally alkaline seawater very slightly more acidic. The effect is hard to measure, but can be predicted with some certainty. As this "CO^-labelled" water sinks to the ocean floor in polar regions and becomes entrained in the deep ocean flows, it will eventually encounter a boundary where the calcium carbonate on the ocean floor is very sensitive to such chemical attack. The carbonate layers on the ocean floor will begin to dissolve. Since this signal will remain locked in the deep sea for hundreds of years, it is not an urgent human concern, but is a fascinating scientific phenomenon. Compared with the total quantity of buried remains of organisms, the fossil fuel reservoir on which our modern industrial economy depends is a small subcomponent. It is, however, large compared to the reservoir of carbon dioxide in the atmosphere. The human acceleration of the rate at which fossil fuel carbon is burned and returned to the atmosphere is generally accepted as the major cause of the well-documented increase in atmospheric carbon dioxide content during the last century. Only about half of all the CO2 that has been produced by the burning of fossil fuels now remains in the atmosphere. The CO2 "missing" from the atmosphere is the subject of an important debate — since the pathways and fates are at the heart of the questions about the damage man is doing to the environment. The great majority of the missing CO2 has invaded the ocean, for this system naturally acts as a giant chemical regulator of the atmosphere. Those engaged in the debate over the details of this process — atmospheric, terrestrial, and ocean scientists — argue over whether the ocean has taken up 40 percent or 50 percent of the emitted CO2. One vexing problem is that the changes are very hard to measure: we can calculate a theoretical result; we can provide "proof" by measuring tracers of carbon — and other chemicals also added to our world by man; but the challenge of producing a record of the changing ocean to match that of the atmosphere still lies ahead. It is clear, however, that ocean processes have a major role in the regulation of the carbon dioxide content of the Earth's atmosphere. From analysis of gas bubbles trapped in ancient ice, it is apparent that the carbon dioxide concentration has varied by about a factor of two during the last few hundred thousand years, and that this cycle is highly correlated with the ice age cycle of the Quaternary Period (from 2 million to 10,000 years ago). To provide quantitative answers to questions relating to the ocean's role in the natural cycle of climate, and to know how ocean properties and processes might change in response to anthropogenic modification of this cycle, observations and models must focus on key aspects of the ocean's biogeochemistry. Much consideration has been given to this. Marine Primary Productivity A major uncertainty in global ocean data is the rate of photosynthetic activity, or primary production, for marine phytoplankton. Although these plants are capable of population doublings every few days, loss terms, such as grazing by zooplankton and sinking deeper than the sunlit layer, typically prevent the size of the plankton population from increasing perceptibly over this period. Since it is usually not possible to determine the rates of production with time series measurements of population size, the alternatives are to measure either the rate at which carbon dioxide is consumed or the rate at which oxygen is produced. The former became possible with the introduction of the carbon-14 tracer technique in plankton production studies early in the 1950s, and its use in the following two decades permitted the first comprehensive global estimates of the rate for marine primary production (Figure 1). This general global view of marine primary productivity is probably reasonably correct, but we now believe that a more accurate representation is both needed and possible. It is evident that the highest rates of production occur in regions that are periodically cooled and enriched by vertical mixing which brings nutrients from the deep ocean to the surface. There is an abundance of carbon dioxide dissolved in seawater, and except at very high latitudes, it is the availability of other nutrients, particularly nitrate and phosphate, that limits the rate of primary production in the sun-lit surface waters. FHigh latitude phytoplankton can grow at temperatures close to the freezing point of seawater, and in polar regions, it is light that seasonally exerts the most effective control on production processes. At temperate latitudes. 20 Figure 1. Distribution of primary plant production in the world ocean, average values per square meter of ocean surface per year. (From Broecker and Peng (1982), adapted from a map in Koblentz-Mishke and others, 1970) primary production has an annual cycle, which is intluenced both by light and by seasonal convection or storm-induced mixing of nutrients upward from depths of several tens to a few hundred meters. The combined effects of winds and ocean currents result in pronounced upwelling of some eastern boundary currents, along the equator, and in the Southern Ocean, and this upwelling of nutrient rich water also stimulates plankton growth. In general then, both the seasonal mixing in temperate regions and the upwelling at particular localities over a wide range of latitudes are responsible for stimulating production and giving rise to a similarity in patterns for cool surface ocean temperatures and primary production. With time these upwelled waters warni, phytoplankton consume the nutrients, and, as the nutrients become depleted, the rates of primary production decline. Techniques suitable for assessing rates of primary production from changes in oxygen concentration with high precision, and others for tracing the photosynthetic production of oxygen with isotopic techniques are available, but have yet to receive wide application. Indirect approaches involving oxygen have, however, had a significant impact on research related to plankton production. In some ocean regions, the oxygen produced by plankton can be trapped below the immediate warm surface waters for several months by a strong density gradient. The annual cycle is completed when strong winter-time mixing "ventilates" the water column to the depth of the main thermocline and permits accumulated oxygen to escape to the atmosphere. Estimates for rates of primary production necessary to create these seasonal surpluses of oxygen seem to be higher than those typically arising from studies using the carbon-14 tracer technique, forcing ocean scientists to struggle to balance their chemical budgets. An independent assessment of the rate of primary production can be attained from the rate at which organic material sinks from the photosynthetic region, or euphotic zone, of the water column. Containers known as sediment traps can be suspended at fixed depths to collect sinking particles, which consist of intact phytoplankton cells and remains of plants and animals. These data can be used in conjunction with measurements of the respiratory consumption of organic matter within the waters shallower than the trap in order to estimate the rate of primary production. Within the last few years, considerable progress has been made in reconciling disparate estimates of marine primary production. Notable among these are improved applications of the carbon-14 techniques, which now yield higher estimates of production than have been reported in the past. Moreover, investigators have begun to 21 combine multiple approaches in studies addressing the rates ot plankton production and consumption, and success with these has contributed to the consensus that oceanographers are now ready to address these issues on a global scale. However, the technology to accomplish this is not yet in place. The need, at this point in time, is for scientists to design and build sensors and instruments to measure these processes. It represents a difficult technical challenge. While the highest rates of primary production occur in coastal regions, and these systems can be studied effectively with ships and moored arrays, the most challenging region is the oceanic province lying beyond the continental shelves. Although the brilliant blue waters of the open ocean are often thought to be relatively unproductive, on a per area and per year basis, their rates of primary production average about half those for coastal waters and about a third those for upwelling regions. In part, this is because the coastal regions have greater amplitude in the annual cycle of nutrient supply and primary production, giving rise to seasonal "plankton blooms." However, as a consequence of the great areal extent of the oceanic province, which amounts to 90 percent of the total ocean area, as much as 80 percent of the oceans' primary production occurs in these central ocean regions. Details regarding both the spatial and temporal distribution of primary production in oceanic waters remain poorly known. Whereas global computations include substantial data sets for some regions of the Northern Hemisphere ocean basins, coverage for the more extensive oceanic regions of the Southern Hemisphere is very sparse. Moreover, data for a complete annual cycle are rare even for the northern regions. One notable demonstration of the seasonal variability in oceanic production is a three-year time series in the Sargasso Sea near the island of Bermuda (Figure 2). Although a brief and intense seasonal bloom is evident in these data, its magnitude is variable over the period of the study. Such plankton blooms, whether in coastal or oceanic regions take on particular significance in the study of marine biogeochemical cycles. These Figure 2. Gross primary production at Station 'S' in the Sargasso Sea off Bermuda (After Menzel and Ryther, 1961) represent periods when plants are produced at rates in excess of herbivores' ability to graze. In the last decade, it has become evident that the flux of biogenic particles to the deep sea is also highly seasonal. This confirmed the supposition that seasonal blooms in the open ocean, although poorly documented, are common and that these blooms are important in terms of export of biogenic particles, including associated skeletal materials to the deep sea. The study of plankton blooms is hindered by their ephemeral nature — they can peak and dissipate in a few weeks time — and by the near impossibility for ship-bound oceanographers to sample synoptically for other than very small regions. The Role for Satellites This temporal and spatial variability is a regular feature of marine ecosystems and occurs over a broad spectrum of time and space scales. To span the range from kilometer scale to global features, and from the short-term, ephemeral events to the interannual, global climate and circulation driven changes in the ocean biota, is the role of satellite remote sensing. Whereas local studies are essential for refining our understanding of certain critical physical/chemical/biological processes, satellite observations provide the best opportunity for extrapolating these local observations to regional, basin, or global scales. For a study of the biogeochemistry of the oceans from space, the changes in ocean color that can be detected by a sensor such as the Coastal Zone Color Scanner (CZCS) have been shown to provide a quantitative measure of near-surface phytoplankton pigment concentrations. These concentrations, which for remote sensing applications represent the sum of chlorophyll-a and phaeophytin-a, are an index of phytoplankton biomass and may be empirically related to primary production. In addition to providing information about the distribution and abundances of phytoplankton, one of the most important features of these measurements lies in their role as a link between the major physical and chemical processes that take place in the ocean and the first step in the system of biological production. As stated in the National Research Council report entitled "A Strategy for Earth Science from Space in the 1980s and 1990s," the first priority is to measure the concentration of chlorophyll-a in the world's oceans. This is also the priority of a CZCS processing effort being undertaken at NASA's Goddard Space Flight Center. The major objective of this program is to produce global scale maps of the time and space distribution of phytoplankton biomass and primary productivity in the world's oceans. To do this, approximately 65,000 individual 2-minute CZCS scenes, each covering an area roughly 1,500 by 800 kilometers, which were acquired between November 1978 and June 1986, will be processed over the next few years. This global data set will allow us to begin to address more effectively questions concerning the interrelationships among climate, the oceans, and their biology. 22 Figure 3. Satellite ocean color image showing the distribution of phytoplankton pigments in the North Atlantic Ocean. This computer-generated image, color-coded according to concentration range, is the first time and space composite of phytoplankton distributions and abundances for an entire ocean basin. This image was produced from data collected during May 1979 using the Coastal Zone Color Scanner (CZCS) aboard the NASA Nimbus-7 satellite. Regions of high pigment concentrations, which are colored yellow and red in this image, represent not only areas of increased phytoplankton biomass, but also reflect periods of enhanced phytoplankton production. The major features of interest include a pronounced band, rich in phytoplankton, across the entire North Atlantic. This spring "bloom" is seen for the first time as a coherent feature across the entire basin. Also evident in the image are the localized regions of high productivity, such as the North Sea, the productive regions along the ice edge, the coastal upwelling zones along the coasts of northwest Africa and South America, and the outflows of the Amazon, Orinoco, and Mississippi Rivers. The number of days in the composite vary from 0 (black area) to 14. White indicates cloud cover. Before the full-scale processing effort begins (lanuary, 1987), a pilot system was established so that the procedures, methodologies, and products could be refined. This pilot system produced the first space/time composite of phytoplankton concentrations for an entire ocean basin (Figure 3). The North Atlantic ocean basin was selected as the test case during May 1979 because of the density of CZCS coverage. A total of 450 CZCS scenes were processed, and the resulting satellite-derived chlorophyll images were remapped into the predefined North Atlantic sampling grid (100W-10E Longitude, 80N- 20S Latitude) to produce 31 individual daily 23 TTO SURFACE (1-15M) LONOTDDEIDEGI -75 -65 -55 -45 -36 -25 -6 jJJ LATTRJDE (DEG) 75 65 55 45 35 25 '^^^DE (DEQ,^^ Figure 4. The partial pressure of CO2 gas in surface seawater expressed as a departure from atmospheric equilibrium. Units are parts per million in volume terms, expressed as microatmospheres. Negative values, or "holes" imply a CO2 flux from the atmosphere to the ocean, and "peaks" imply a CO2 flux from the ocean to the atmosphere. (Courtesy Dr. Peter Brev\/er, WHO!) mosaics, which showed the CZCS coverage for each particular day. These mosaics were then composited over weekly and monthly time scales. Each picture element (pixel) in the composited images represents the average chlorophyll concentration within a 24 x 24 kilometer area of ocean surface. The number of pigment retrievals at each sampling point varied, with the densest sampling along the coastal margins and the fewest number of points in the open ocean. In fact, the large black region in the center of the monthly composite shows that no CZCS data were collected at all over this area during May 1979. Although significant mesoscale variability was observed over short time scales (daily to weekly), monthly CZCS composites appear to retain the major mesoscale structures and dominant features of the region and will be the best means for quantifying the large-scale, interannual variability in global ocean primary production. Global Ocean Flux Study In the last two years, groups of oceanographers from several countries have begun to define a program known as the Global Ocean Flux Study. Its primary aim is to observe and understand the biogeochemical cycles of the ocean sufficently well to predict the interaction between the oceanic, atmospheric, and sedimentary cycle for carbon and associated elements, such as nitrogen, oxygen, and sulfur. It is envisioned that this program will be phased with other major ocean programs that are scheduled for early in the 1990s as part of the new National Science Foundation focus in Global Geosciences. For one example of the value of such a program, we refer the reader to Figure 1, page 9, of the Moore and Bolin article showing the atmospheric CO2 record. Carbon dioxide in the atmosphere has no active chemistry; that is, it is not created or destroyed by any chemical reactions whatsoever. It is simply mixed around the globe by winds. The peaks and valleys in this record result from the atmosphere being fed by waves of CO2 put into, and pulled out of, the air by land and sea. The land fluxes are more rapid and produce the dominant short-term signals. The ocean fluxes are smaller (Figure 4), but have immense capacity. An ability to observe and predict these changing fluxes would be of enormous value to those who are concerned with global change. 24 The World Ocean Circulation Experiment /\n ambitious international program called the World Ocean Circulation Experiment (WOCE) — probably the most complex experiment that the world oceanography community has ever attempted — is scheduled to begin late in 1987. It will provide by 1 995 the first comprehensive global survey of the physical properties of the oceans, information vitally needed to understand global climate dynamics and to predict climate change on a scale of decades. The results also will provide the impetus for new research in marine chemistry, biology, and geology. WOCE Is being coordinated under the World Climate Research Programme by the Scientific Committee on Oceanic Research, the World Meteorological Organization, and the Intergovernmental Oceanographic Commission. U.S. participation involves several new oceanic initiatives and the expansion of a number of ongoing activities. ,^mong the new elements required for the implementation of this experiment are a series of new ocean operational and research satellites. These include the Geodetic Satellite (GEOSAT), the Navy Remote Ocean Sensing System satellite (NROSS), the Earth Resources satellite (ERS-1), the Ocean Topography Experiment satellite (TOPEX), and the Geopotentlal Research Mission satellite (GRM). Another element involves the establishment of a global density and tracer program, which includes the completion of a global description of transient tracer distributions and a resurvey of the density field of the oceans. This work will require one or more dedicated vessels. A global program of circulation velocity mea- surements using ocean drifters and floats are of central importance to the overall experiment. Such a program is under consideration. Another critical need is the development of a new sys- tem to manage data, not only that provided by WOCE, but for other experiments as well such as the Tropical Ocean and Global Atmosphere Program (TOGA). A coherent program of sea-level measure- ments is also planned to enhance the irregular network of such measurements now in place and contributed to by many countries. Measure- ments from sea-level gauges located by the Global Positioning System satellite and very long baseline interferometry will be used to de- fine a global data set and to verify initial satellite altimetric measurements of ocean topography. Many other programs are planned under the WOCE umbrella. Only a few can be mentioned here. The U.S. Science Steering Committee for WOCE has defined its primary scientific objec- tive as being: "to understand the general circu- lation of the global ocean well enough to be able to model its present state and predict its evolution in relation to long-term changes in the atmosphere." Specific Objectives are: • To complete a basic description of the general circulation of the ocean. • To determine seasonal and interannual oceanic variability on a global scale and the effect of such variability on ocean measurement strategies and the co-evo- lution with the atmosphere. • To improve the basic description of the surface boundary conditions and the ex- changes of physical properties with the atmosphere, and to establish their un- certainties. • To determine the interbasin exchanges in the global ocean circulation. • To obtain quantitative estimates of the large-scale exchange of buoyancy and chemical constituents between the up- per boundary layer and the ocean inte- rior, by adequately describing the prop- erties of the surface layer, including its horizontal mass transport and diver- gence. • To determine oceanic heat transport and storage in relation to the heat budget of the earth. • To determine the large-scale transport capacity for ideal tracers in the ocean. • To determine the important processes and balances for the dynamics of the general circulation. • To improve numerical models for the diagnosis, simulation, and prediction of the general circulation of the ocean. Readers wishing to know more about this project should write: U.S. Planning Office for WOCE, Department of Oceanography, Texas A&M University, College Station, Texas, 77843. The U.S. WOCE Science Steering Committee in- cludes: D. lames Baker, jr. (joint Oceanographic Institutions, Inc.), Francis Bretherton (National Center for Atmospheric Research), Dudley Chel- ton (Oregon State University), Russ Davis (Scripps Institution of Oceanography) William Jenkins and Terrence Joyce (Woods Elole Oceanographic Institution), William G. Large and lames C. McWilliams (NCAR), Worth D. Nowlin, jr. (Texas A&M University), Ferris Webs- ter (University of Delaware), Ray Weiss (Scripps), and Carl Wunsch (Massachusetts Institute of Technology). 25 A change in climate can be expected to alter the productivity and food chain relationships of the world's ocean. (Wilson North © ; 985) The key elements of the Global Flux Study are: 1). The use of satellite sensed ocean color data to estimate plankton distribution on the necessary space and time scales; 2) The correlation of this signal with direct observations of upper ocean chemistry and biology; and 3) Measurements of the rain rate of organisms and their remains to the ocean floor and its relationship to the sediment accumulation flux. Essential complementary data on the physical template of ocean heating, cooling, mixing, and transport processes will be provided by programs like the World Ocean Circulation Experiment (WOCE), described in the box on page 25, and the Tropical Ocean Global Atmosphere Program (TOGA). The translation of these observations into the first comprehensive global view of ocean primary productivity, carbon flux to the deep ocean, and other variables of critical importance to our understanding of the coupling of marine biogeochemical cycles and the physical climate system will involve a novel mix of measurements and models on the largest scale. A change in climate like that predicted for the next century in response to increasing atmospheric concentrations of CO2 and other greenhouse gases can be expected to alter the upper ocean conditions that influence oceanic plankton blooms. Because of the enhanced downward flux of particulate carbon associated with blooms, a change in bloom conditions can have feedback to climate. The time scale for this may, in general, be long, but recent models have shown a particular sensitivity in high latitude regions where this feedback may have global consequences on much shorter time scales. James /. McCarthy is Director of the Museum of Comparative Zoology and Professor of Biological Oceanography at Harvard University. Peter C. Brewer is a Senior Scientist in the Chemistry Department at the Woods Hole Oceanographic Institution. Gene Feldman is a scientist at the Coddard Space Flight Center of the National Aeronautics and Space Administration. 26 The Oceans as a Source of Biogenic Gases by Meinrat O. Andreae I he atmosphere is part of the global life support system: it supplies oxygen to us and all other organisms (with the exception of some types of bacteria) and it removes our gaseous waste products, especially carbon dioxide (CO2). It protects us from lethal components of solar and cosmic radiation. It redistributes the sun's heat around the Earth, and, in the process, creates weather and climate. The functioning of this life support system is dependent on the maintenance of relatively narrow tolerances in the composition of the atmosphere: for example, too much COj, and temperatures worldwide would rise to make large regions unsuitable for life; too much oxygen, and forests would ignite spontaneously; too little ozone in the stratosphere, and solar ultraviolet radiation would increase, and with it, the incidence of skin cancer. It is not surprising, then, that the biosphere, defined here as the ensemble of all living organisms, actively "maintains" the atmosphere in a composition that is favorable to living organisms. Were it not able to do so, conditions at the Earth's surface would proceed towards chemical equilibrium and might resemble the climate of Venus or Mars. This close relationship between the existence of life on a planet and the composition of its atmosphere was recognized two decades ago by James E. Lovelock, a private researcher who lives and works at Coombe Mill, Cornwall, England. From this idea grew the "Gaia" hypothesis (named after the Greek word for Earth), a still controversial concept which postulates that Earth is a self- regulating system comprising both the living organisms and their environment. This system is suggested to have the capacity to maintain the global climate and chemical composition of the atmosphere at a steady state favorable to life. Many of the major and trace components which make up the Earth's atmosphere are the result Above, the author with air sampling equipment on the mast platform of the R/V Columbus Iselin (University of Miami). 27 ATMOSPHERIC OXYGEN CONTINENTAL ROCKS ! MOVEMENT OCEAN /ATMOSPHERE FLUX DISSOLVED OXYGEN IN THE OCEANS MARINE BIOTA OF EARTH CRUST jORGANIC CARBON ^ SEDIMENTATION MARINE SEDIMENTS Figure 1. The global biogeochemical oxygen cycle. Solid arrows represent the flux of oxygen (heavier arrows indicate larger fluxes). The broken line from the marine biota to the marine sediments and continental rocks shows the flow of organic carbon, which represents an "oxygen demand." (Diagram simplified) of the emission of volatile substances by biological organisms — microbes, algae, plants, and animals. These "biogenic" emissions are such a common phenomenon that we have evolved a very sensitive and selective organ to detect and characterize these emissions: our nose. At the seashore, for example, we notice a variety of smells, some of which may be quite intense at times, for example, the rotten-egg odor of hydrogen sulfide (H.S) bubbling out of intertidal sediments, or the smell of decaying fish, which is caused by the release of volatile amines (nitrogen compounds). And then there is the much fainter, but very characteristic "odor of the sea," which is mostly due to the sulfur compound dimethylsulfide (DMS). Among the biogenic gases which escape from the oceans into the atmosphere, a few compounds are currently considered to be the most important to the chemical and physical characteristics of the atmosphere: oxygen, the sulfur compounds dimethyl sulfide (DMS) and carbonyl sulfide (COS), and the methyl-halogen compounds methylchloride, methylbromide, and methyliodide. The Global Oxygen Cycle Our atmosphere is set apart from that of the other planets by the presence of large amounts of oxygen, about 20 percent by volume. The oceans play a central role in regulating the cycle of atmospheric oxygen. The role of oxygen in the global biogeochemical cycle and in the evolution of the atmosphere and oceans are being studied by Robert M. Carrels of the University of South Florida and by Heinrich D. Holland of Harvard University, together with their coworkers. Oxygen is produced by plants and algae during photosynthesis, the conversion of CO2 and water to organic matter and oxygen. The organic (carbon-containing) compounds produced by plants and algae and the solar energy stored in these compounds are the basis for food chains on land and in the sea: the biochemical "combustion" of organic compounds, especially sugars and other carbohydrates, is the energy source for all organisms, including ourselves. This process — called respiration — is the reverse of photosynthesis. It consists of the reaction of oxygen with organic matter, to produce CO2 and water (Figure 1). On land, nearly all of the organic matter produced is rapidly consumed again by respiration. Rapidly, that is, on a geological time scale: most of the organic matter is reoxidized in months or years, almost none of it persists longer than a few hundred years. Since the consumption of organic matter by respiration requires a corresponding amount of oxygen, the production and consumption of oxygen by the land biota are roughly balanced, and no net input of oxygen into the atmosphere takes place on the continents. In the oceans, however, unlike on the land, there is an imbalance between the production and the consumption of oxygen, due largely to the process of marine sedimentation. At the bottom of the oceans, organic matter is incorporated into the marine sediments. While this buried organic matter is only a small fraction (on the order of 1 percent) of the total amount produced by marine photosynthesis, it does lead to a long-term removal of organically-bound carbon from the surface of the Earth. Since oxygen was produced by photosynthesis when this organically-bound carbon was formed, and since this oxygen is not consumed again by respiration in the sea, it will escape into the atmosphere from the ocean's surface, where the exchange of gases between ocean and atmosphere takes place. The annual amount of oxygen that is released by this process is on the order of 300 million metric tons. If there were no oxygen loss from the atmosphere, this input would lead to a doubling of the atmospheric oxygen content in about 4 million years, a short time geologically. Where then does this oxygen go? The answer lies back with the marine sediments. Marine sediments containing organic carbon are transformed into rocks like shales and schists by geological processes in the Earth's crust. These rocks are eventually uplifted on land and undergo erosion and weathering. Oxidation of the reduced constituents of these rocks — ferric iron, sulfides, and organically-bound carbon — consumes atmospheric oxygen, and this process balances the atmosphere's oxygen budget. The word "budget" is used here in a geochemical sense: for a given substance or chemical element, we compare the sum of all the inputs to the atmosphere with the sum of all the outputs from the atmosphere. If inputs and outputs are of the same size, the budget of the substance is "balanced," and its abundance in the atmosphere remains the same over time. If they are not balanced, there is either some source or removal mechanism that we have not estimated correctly, or the amount of the substance in the atmosphere is 28 changing, just as a deficit in a financial budget draws down the balance. Since the biosphere controls the rate of oxygen production in the sea, and since the rate of oxygen consumption on land depends on the rate of weathering, the input and removal of oxygen to and from the atmosphere may not be linked by a common process. In the absence of such a linkage, the atmospheric oxygen levels could undergo wide swings, depending on which process happens to dominate at a given time. The feedback mechanisms, which provide this linkage and thereby keep atmospheric oxygen within tolerable limits, are still poorly understood. Present theories suggest that feedback takes place through an adjustment of the proportion of marine organic matter which escapes reoxidation in the sea and is buried in sediments. If the oxygen content of the atmosphere should drop because weathering consumes more oxygen than the excess oceanic photosynthesis provides, the oxygen concentration in the oceans would also drop. As a consequence, the rate of oxidation of organic matter in the oceans would fall, and the amount of buried carbon would increase. This, in turn, would increase the amount of photosynthetic oxygen that is not consumed in the ocean, thereby increasing the oxygen level in the sea and, subsequently, in the atmosphere. This process therefore provides a negative, stabilizing feedback, protecting the atmosphere from wide swings in oxygen content. Marine Biogenic Sulfur Emissions While the air/sea exchange of oxygen, a major atmospheric constituent, is of central importance in life processes, and is crucial to the life support function of the atmosphere, most of the volatile substances that are produced by life in the sea are emitted only in relatively small amounts. Small amounts in a geochemical sense, of course, are still measured in millions of tons! The ocean emission of such trace constituents results in atmospheric concentrations that are on the order of parts per trillion to parts per billion. Do such trace levels still have global significance? Surprisingly, global climate, a process which involves the transport of gigantic amounts of energy and mass, may be quite sensitive to changes in some of the trace constituents in the atmosphere. The role of CO2, which is present at the part per million level, is discussed in the article by Moore and Bolin in this issue (page 9). Here, we shall look at a trace sulfur compound, dimethylsulfide (DMS). Its emission from the seas may contribute to preventing an overheating of the Earth's surface and the atmosphere. The role of DMS in the global atmospheric sulfur cycle has been studied during the last few years by myself and my colleagues at Florida State University, and by Peter Liss and Susan Turner at East Anglia University, Norwich, England, as well as at a number of other institutions in Europe and the United States. The presence of sulfur compounds in the atmosphere has been documented since the 19th century, when sulfate levels in rainwater were systematically measured in England. Yet, when a sulfur budget of the atmosphere was calculated based on the estimates of known sulfur emissions on one hand, and the measured rates of sulfate deposition on the other, more sulfur seemed to come out of the atmosphere than could be accounted for by known inputs. With the odor of hydrogen sulfide, H7S, from marine muds on their minds, early investigators postulated that this deficit in the atmospheric sulfur budget was explained by marine emissions of HiS. This hypothesis was disproven, however, since the same environmental circumstances that lead to the production of hydrogen sulfide also prevent its escape to the atmosphere. Hydrogen sulfide is the product of an alternative mode of respiration, which only occurs in the absence of oxygen, and which uses sulfate instead of oxygen to oxidize organic matter. It therefore is abundant only in environments in which the supply of oxygen by exchange with the atmosphere is cut off for some reason — for example, in the porewater of sediments. The same circumstances, however, which keep the oxygen out, also keep H2S in. Subsequently, measurements of HjS in the marine atmosphere showed concentrations that were much too low to explain the deficit in the atmospheric sulfur balance. Dimethylsulfide (DMS), on the other hand, is produced by algae in the upper 50 meters or so of the oceans, the layer which receives enough light for the growth of photosynthetic organisms (Figure 2, page 32). Gas exchange between this layer and the atmosphere is rapid, on the order of days to weeks. While the concentrations of DMS in the surface ocean are quite low, on the order of billionths of grams per liter of seawater, DMS is present everywhere at the sea surface, and even a small flux per unit area, multiplied by the tremendous surface area of the ocean, provides a sizeable input into the atmosphere. Our best current estimate of the magnitude of the input of DMS from the oceans into the atmosphere is about 60 million tons per year, or about 30 million tons of sulfur. This corresponds to about a third to a half of the sulfur that enters the atmosphere from fossil fuel burning. But, whereas the input from fossil fuel combustion is centered on the industrialized regions of the Northern Hemisphere, the DMS flux is distributed almost evenly over the oceans worldwide. It therefore represents most of the sulfur input into the remote marine atmosphere, especially in the Southern Hemisphere. Once in the atmosphere, DMS is rapidly (on the order of days) oxidized, with sulfuric acid being one of the main products. Sulfuric acid is quite involatile and condenses into small particles. Caseous ammonia also is incorporated into these particles. The result is a fine aerosol, consisting of a mixture of ammonium sulfate and sulfuric acid, with a particle diameter of a fraction of a millionth of a meter. Eventually, these particles are incorporated into rain and return to the oceans or the land surface. While the amount of sulfuric acid which this natural source contributes to rain produces some acidity, it is far below values that have been 29 Ozone il igh in the gaseous envelope surrounding the Earth is a wavy layer containing a small but vital quantity of the oxygen molecule O3, or ozone (some .001 percent). Ozone is constantly created by the action of sunlight on normal oxygen molecules, and is as constantly destroyed after about 18 months through interaction with nitric oxide and other molecules rising from the earth. During that time it drifts from the equator towards the poles, where the ozone layer is at its thickest. It is subject to considerable natural fluctuation, and responds to changes in solar activity. Over the 22-year sunspot cycle it may vary by as much as 12 percent. Ozone has two closely related functions. First, it acts as a buffer to much of the ultraviolet radiation from the Sun (in particular, the wavelength of 0.26 microns, which would otherwise damage the DNA or reproductive molecule in all living systems). As a rule of thumb, ultraviolet radiation reaching the ground increases about 2 percent for each I percent decrease in atmospheric ozone. Variations in the thickness of the ozone layer are reflected in the ability of organisms, including the human species, to cope with different degrees of ultraviolet radiation. Thus, white-skinned people who venture into the tropics or climb to high altitudes are more subject to sunburn and skin cancer than brown- or black-skinned inhabitants. Secondly, by absorbing ultraviolet radiation the ozone layer heats the stratosphere, causing the familiar problem of heat inversion as observed so often at a much lower level over such cities as Mexico and Los Angeles. Ozone thus forms a kind of lid on the atmosphere, and a rich variety of natural and man-made products gets trapped beneath it. Some may do it active harm. The most notorious are the chlorofluorocarbons (known commercially by a variety of names, including freons), which are used as a propellant in spray or aerosol cans and as a refrigerant in cooling devices. Chlorofluorocarbons have a similar effect to carbon dioxide in blocking some infrared radiation from the Earth, and thus retaining heat at its surface. They rise slowly into the stratosphere, where they are turned by the actions of sunlight into fluorine and chlorine atoms, some of which destroy ozone. The chemical oxides of nitrogen are less efficient but some again have a similar effect: here the most important are those injected into the stratosphere from the exhausts of high-flying aircraft, from nuclear explosions, or from chemical processes on Earth. Nor is this all. Other suspects are the nitrous oxide produced in nitrogen fertilizer, brominated and chlorinated compounds used in the purification of drinking water and sewage, and carbon monoxide from car and other exhausts. There is as yet little certitude about the effects of human activity on the ozone layer. With wide normal fluctuations, the human contribution is hard to determine. The level of ozone over Antarctica as reported in October of 1 986 has dropped precipitously since 1978, with the rate of decline especially rapid since 1982. The drop begins in late August or early September as sunlight returns to the polar region, following the end of polar night. It has averaged as much associated with the problem of "acid rain" caused by emissions from the burning of fossil fuels. During the few days these sulfate particles spend in the atmosphere, they interact with sunlight to exert their influence on the Earth's climate. The temperature of the Earth's surface is the result of the balance between the incoming radiation from the sun and the outgoing radiation emitted by the Earth (Figure 3, page 32). The incoming radiation is dominated by light waves of relatively short wavelength, the visible part of the spectrum. Some of this incoming radiation is reflected back into space by clouds, air molecules, dust particles, and the Earth's surface. The rest is absorbed at the surface and heats the Earth. Just like a warm stove, the Earth gives off its heat energy in the form of infrared, or long-wave, radiation. The intensity of this radiation increases with the surface temperature of the Earth; therefore this temperature will assume a value at which the outgoing long-wave energy exactly balances the incoming short-wave energy. At this temperature, the Earth is in radiative equilibrium. Consequently, the surface temperature of the Earth can be altered either by changing the efficiency with which incoming energy is absorbed, or by adjusting the degree to which outgoing radiation is permitted to escape. Global warming by the increased retention of outgoing long-wave radiation is called the "greenhouse effect." This effect is expected to lead to a significant increase in temperatures worldwide in the next few decades because of the buildup of COj from the burning of fossil fuels. While there is an intensive scientific discussion going on about the CO^-induced greenhouse effect, it is often forgotten that other gases with similar optical properties are also 30 Ozone over the Southern Hemisphere, as mapped on October 16, 1986, by the total ozone mapping spectrometer (TOMS) aboard the Nimbus-7 satellite. The ozone concentration over the Antarctic is composed of a "hole" ot \ow ozone values (purple area in center), ringed by a band of higher values, set into a background of more typical concentrations over the rest of the hemisphere. The year-to-year deepening of the ozone hole is presently without any simple or universal explanation. (Image courtesy A. J. Krueger, Goddard Space Flight Center, NASA) as 1 percent a day in September in recent years. Removal of O^ is particularly significant at low altitudes, between 10 and 20 kilometers. It is obviously important that we develop a rapid understanding of what has come to be known as the ozone hole. Three theories have been advanced. One argues that removal of O^ may be attributed to enhanced concentrations of nitric oxide associated with increased ionization of atmospheric gases during periods of high solar activity. A second suggests that the drop in O3 is caused by dynamical influences, vertical motion carrying tropospheric air with low O3 into the stratosphere. The third attributes the fall in O3 to chemical reactions involving halogen radicals, with an important contribution due to carbon 10 formed as a consequence of the decomposition of manmade chlorofluorocarbons. The chemical theory suggests that levels of nitrous oxide, NO2, should be very small, in contrast to the solar activity model which requires that they be large. Measurements made this year by a group of scientists at McMurdo station in Antarctica suggest that levels of nitrous oxide in the ozone hole are very low. This would appear to eliminate the solar activity theory, heightening the prospect that human activity may be implicated in removal of Antarctic O3. The ozone hole — a small one has also been noted over the North Pole — is the subject of vigorous research at the moment. It is hoped that measurements can be carried out in 1987 with instrumentation on high-flying aircraft. It is important that we develop rapidly an understanding of the phenomenon in order to assess its implications for the larger global environment. —from C. Tickeli 1986, Climatic Change and World Affairs, and from material supplied by Michael McElroy of Harvard University. increasing in the atmosphere. Of interest here are methane and nitrous oxide, both of which are emitted from the oceans, although at rates that are low compared to other sources. The sulfate aerosol, on the other hand, influences the amount of incoming short-wave radiation reflected back into space before it can be absorbed at the Earth's surface. This can happen either by direct scattering of light on the sulfate aerosol particles, or by the influence of the aerosol on the properties of clouds. The particle size of the sulfate aerosol is approximately the same as the wavelength of light; it turns out that this is the most effective particle size for scattering and reflecting light. However, it appears that the concentration of sulfate aerosol in the atmosphere would have to increase substantially over present natural levels to have a significant influence. Such an increase is seen in regions influenced by human emissions or by volcanic eruptions, but probably exceeds the range of variability of biogenic emissions. There may, however, be an effect which amplifies the climatic influence of sulfate aerosol particles over the ocean by interaction with the processes which lead to the formation of clouds, as has been proposed recently by Lovelock, Robert Charlson of the University of Washington, and myself. In order for cloud droplets to form, air must be supersaturated with water vapor. This usually happens in the atmosphere when an air parcel rises and consequently cools down. The droplets then form by the condensation of water vapor on existing particles (called cloud condensation nuclei). The abundance of nuclei determines the number of cloud droplets formed in a given volume of air. Over the remote oceans, where little 31 Carbonyl sulfide Sulfur dioxide Sulfate STRATOSPHERE TROPOSPHERE Sun \ \ light \ \ \ \ ATMOSPHERE _^ OCEAN M ^ Carbonyl sulfide Dissolved ^ organic X sulfur compounds Sulfate Sulfur dioxide Dimethylsulfide Precipitation Sulfate Volatilization Dimethylsulfide [Planktonic algae[^ Figure 2. The cycle of biogenic sulfur compounds emitted from the ocean to the atmosphere. Dimethylsulfide is produced by planktonic algae, escapes into the atmosphere, and is oxidized to sulfate aerosol particles in the troposphere (the lower part of the atmosphere). Carbonyl sulfide is released by the photodecomposition of organic sulfur compounds and oxidized to sulfate in the stratosphere. (Diagram simplified) SPACE Incoming solar radiation 100 7< Outgoing radiation Short-wave Long-wave 67o 387o A 267o ATMOSPHERE Absorbed by water vapor, dust, O3 Absorbed by clouds Reflected by surface / Emission by water vapor, COo Absorption 50/ by water vapor, CO2 Emission by clouds Heat flux by evaporation Surface emission of long-wave radiation Heating of air by ground OCEAN, LAND 5l7o 2l7o 77o 237o Figure 3. The radiation budget of the Earth. Most of the energy from the Sun arrives in the form of short-wave (visible) radiation. Some of it is reflected into space by air molecules, clouds, and the Earth's surface. The rest is absorbed and heats the Earth. The warm Earth emits long-wave (infrared) light, part of which is absorbed in the atmosphere. The retention of infrared radiation by the atmosphere leads to the "greenhouse effect." 32 continental dust penetrates, the only source of cloud condensation nuclei is the formation of seasalt aerosol from seaspray and the production of sulfate aerosol. Seasalt nuclei are relatively rare: only a few of them are present in one cubic centimeter of air over the ocean. Therefore, most cloud droplets actually form on sulfate aerosol particles, which have number concentrations of a few hundred per cubic centimeter over the remote oceans. Consequently, we can say that biogenic sulfur emissions from the oceans are dominant in controlling the density of cloud droplets in marine clouds, and therefore their reflective properties. While it is difficult to estimate what the exact effect of, say, a doubling or halving of the marine DMS flux on climate would be, it appears certain that, in the absence of the biogenic DMS flux from the sea, the Earth would be significantly warmer. Carbonyl Sulfide and Stratospheric Haze Carbonyl sulfide (COS) is also formed in the upper layer of the oceans. While DMS is produced by the life processes of planktonic algae, COS is formed by a photochemical process, which does not involve living organisms. Of course, this is not quite correct, since the photochemical reaction requires as starting materials both oxygen and organic sulfur compounds, which are formed biologically. It does not seem to matter much which organic sulfur compounds are present: we exposed to light solutions of a number of biologically relevant compounds in seawater, and all of them yielded COS, although at different rates. Because the production of COS in seawater is photochemical, a systematic variation of its concentration in the surface ocean as a function of the time of day results, with the highest values observed shortly after midday and the lowest values at night (Figure 4). NORTH ATLANTIC 26 APR - 10 MAY 1984 1000 i::^ _ 01 03 05 07 09 II 13 15 17 19 21 23 01 LOCAL TIME (h) Figure 4. The concentration of carbonyl sulfide (COS) in the surface waters of the ocean shows a pronounced variation with the time of day. The concentration of COS is expressed here as the degree by which it is supersaturated relative to the overlying atmosphere. The agreement between the daily cycles of light intensity and COS concentration is due to the production of COS by the photochemical destruction of biogenic organic sulfur compounds. Methane I he most abundant hydrocarbon, methane, (CHJ often called natural gas, is increasing in the atmosphere. It is thought to he a natural constituent of the air — arising as it does from many biological processes, and from seepage out of the Earth. Measurements in the 1950s and 1960s were imprecise. This, along with spatial differences obscured any trends. However, in the late 1970s, several investigators using gas chromatography unequivocally demonstrated an upward trend. Increase in the number of farm animals, and hence flatulence, and expansion of rice production might well explain, at least qualitatively, the atmospheric methane growth. Other biological activities, such as termite destruction of wood, possible leakage from man's mining, and use of fossil methane might also contribute to methane in the air, but their contribution to its increase is less clear. The higher concentrations far north of the equatorial region suggest that the termite source may be minor. The relatively rapid recent increase with time, about as fast as for carbon dioxide (CO2), combined with the uncertainty as to its origin, are both intriguing features of the methane growth in air. There is no reason to expect the upward trend in atmospheric methane concentration to stop soon, since the most likely sources of methane are related to population size. In the long run, however, these sources may ultimately be somewhat limited by space. Methane also forms another link in the question of future atmospheric composition. Large amounts of methane are believed to be stored in methane hydrates in continental slope sediments. Methane hydrate is a type of clathrate in which methane and smaller amounts of ethane and other higher hydrocarbons are trapped within a cage of water molecules in the form of ice. Methane hydrates are stable at low temperatures and relatively high pressures. With a rise in ocean-bottom temperatures, the uppermost layers of ocean sediments also would become warmer; and methane hydrates would become unstable in the upper limit of their depth range, that is, about 300 meters in the Arctic and about 600 meters at low latitudes. Some quantity of clathrates may therefore be released from sediments under the seafloor as a result of ocean warming. — from Changing Climate, National Academy Press 33 Nitrous Oxide Mc 'ost nitrous oxide, N2O, in the air has come from denitrification in the natural or cultivated biosphere. The largest part of atmospheric nitrous oxide therefore is derived from nature, unrelated to human activity. Recent, careful measurements have suggested a small growth rate in the concentration of nitrous oxide in ground-level air at remote locations. The source of the small increase is unknown, but prime candidates are the continued expanded use of nitrogen fertilizers around the world to improve agricultural productivity, and high temperature combustion in which atmospheric nitrogen is oxidized. If such fertilizers are the source, then the current slow increase is likely to continue into the foreseeable future because the demand for food will grow with population size. — from Changing Climate, National Academy Press In contrast to DMS, which is rapidly oxidized in the troposphere (the lower 10 to 20 kilometers of the atmosphere, where "weather" takes place), COS is practically inert there. It can only be photo- oxidized in the stratosphere, where energetic ultraviolet radiation is present. There, it also ends up producing sulfate aerosol particles, which form a thin veil of haze in the stratosphere, which reflects sunlight and helps to cool the Earth. Chlorine, Bromine, and Stratospheric Ozone Atmospheric ozone shields the Earth from damaging ultraviolet radiation. While halogen compounds from spray cans and other anthropogenic sources may affect this ozone layer, ozone may also be affected by naturally-produced halogen compounds. A variety of organic compounds of the halogen elements — chlorine, bromine, and iodine — are formed in the sea by processes that are still poorly understood, but which certainly involve living organisms. At our current state of knowledge, the most important of these organohalogen compounds appear to be methylchloride (CH^CI), methylbromide (CH^Br), methyliodide (CH3I) and bromoform (CHBrs). Emission from the sea surface appears to be the dominant source of these compounds to the global atmosphere. The marine and atmospheric chemistry of these compounds have been studied in the last few years by Peter Liss of East Anglia University, England; Hanwant Singh at Stanford Research Institute, Palo Alto, California; and Reinhard Rasmussen and Aslam Khalil at the Oregon Graduate Center, Beaverton, Oregon. Methyliodide is emitted in copious quantities by coastal seaweeds. It also may be released in the open ocean by microscopic planktonic algae, although this has not been demonstrated yet. Its enrichment in the upper layers of the oceans, however, clearly shows that it must be produced in photosynthetically active surface ocean waters. An alternative possibility to its direct production by algae is the reaction of dimethylsulfoniopropionate (DMSP, the precursor of DMS in algae) with iodide ion in seawater. DMSP is released by planktonic algae into seawater, but the rate of the reaction with iodide ion remains to be investigated. The role of methyliodide in the atmosphere remains speculative: it is possible that it may lead by a sequence of photochemical reactions to the catalytic removal of ozone in the marine troposphere. It also has been argued that the iodine- oxygen compound, 10, which is formed photochemically from methyliodide, plays an important, if not dominant role in the atmospheric oxidation of DMS. Much laboratory and field research remains to be done before these possibilities can be evaluated, however. The production mechanisms for methylchloride and methylbromide have not been clearly identified yet, either. The same process as has been suggested for methyliodide, that is, the reaction of DMSP with chloride and bromide, may lead to the formation of methylchloride and methylbromide. Alternatively, it is possible that they are formed from the reaction of methyliodide with chloride and bromide ions in seawater. Our interest in the emission of these compounds from the oceans stems again from their potential role in regulating the chemistry of the stratosphere. Ozone absorbs ultraviolet radiation very effectively; therefore the presence of relatively high levels of ozone in the stratosphere protects us from the damaging effects of ultraviolet light, which include skin cancer. Reactions with chlorine and bromine atoms, which are formed from the photochemical breakdown of methylchloride and methylbromide, remove ozone from the stratosphere. We have become acutely aware of this problem as a consequence of the introduction of the freon compounds (volatile organic compounds containing fluorine and chlorine) as propellants in spray cans and as coolants in refrigerators and air conditioners. The release of the freons into the atmosphere as a result of their widespread use has led to their presence in the stratosphere, where they are broken down to yield chlorine atoms. It is feared that these chlorine atoms then deplete the vital stratospheric ozone layer. The degree to which this ozone depletion is expected to occur is still a matter of active scientific debate. It is clear, however, that we have to clarify the role of natural inputs of chlorine and bromine into the stratosphere if we are to correctly evaluate the effect of disturbances of stratospheric chemistry by human activity. Scientific Problems for the Future Atmospheric chemistry is a young science — the role of the oceans in atmospheric chemistry has only been studied in the last two decades. Some progress has been made in identifying a few of the key processes and compounds which are responsible for the influence of the ocean on the composition and properties of the atmosphere. On the other hand. 34 The compound melhyliodide (CH,/) is emiUed to the atmosphere in large amounts by coastal seaweeds such as the kelp shown here. (Wilson North (a)l985) large groups of compounds of potential importance have received little attention. Atmospheric measurements by lochen Rudolph of the Kernforschungsanlage, Julich, Germany, have suggested recently that reactive hydrocarbons are emitted by the oceans which are important in controlling the photochemical processes in the atmosphere. Various groups have found tentative evidence that organic nitrogen compounds of marine origin could be of importance in the atmospheric nitrogen cycle. These are examples of the diversity of marine compounds of potential relevance to the composition of the atmosphere which may yet have to be discovered. Surprisingly little is known about the biochemical, physiological, and ecological processes that are responsible for the production of important compounds, like DMS and the organohalogen compounds. Since these compounds are produced and consumed by a variety of marine organisms, each with different rates and possibly different pathways, the biological-chemical relationships are highly complex, and it has been exceedingly difficult to make progress in this area. Another crucial problem is the measurement of the rate of transport of substances across the air/ sea interface. At this time, these fluxes are estimated using a simple model that relates the concentration gradient across the air/sea interface to the flux with the help of some poorly defined, empirical relationships. It would be highly desirable, but technically very difficult, to be able to make some direct measurements of this flux. We hope that such measurements will become possible within the next decade at least for some compounds. Once direct flux measurements can be made for some of the "easier" compounds, we will be able to test and improve the flux calculations for those compounds where direct measurements remain impossible. Meinrat O. Andreae is Professor of Oceanography at Florida State University, Tallahassee. Selected References Bucit-Menard, P., ed. 1986. The Role of Air-Sea Exchange in Geochemical Cycling. 549 pp. Dordrecht: Reidel. Holland, H. D. 1978. The Chemistry of the Atmosphere and Oceans. 351 pp. New York: Wiley. Kump, L. R., and R. M. Carrels. 1986. Modeling atmospheric O2 in the global sedimentary redox cycle. American journal of Science 286: 337-360. Liss, P. S., and W. C. N. Slinn, eds. 1983. Air-Sea Exchange of Cases and Particles. 561 pp. Dordrecht: Reidel. Chlorofluorocarbons I his class of gases originates from industrial activities and has been emitted to the atmosphere during the last 50 years. These gases are increasing in the atmosphere approximately as expected from their growth in emissions. CFC-ll, CFC-12, and CFC-22, the three most abundant ones, all have long residence times in the air (tens of years) so that they can accumulate. Both the sources and sinks of the chlorofluorocarbons are believed to be known. The emissions from industrial production and products (such as aerosol propellants) represent the only source of any consequence. Photochemical destruction, mainly in the stratosphere, and very slow uptake by the oceans are the only known significant sinks. Theoretically, chlorofluorocarbons are implicated as potential destroyers of stratospheric ozone, the destruction of which in turn could result in damage to human health and the environment from increased ultraviolet radiation. — from Changing Climate, National Academy Press 35 Man's Great Geophysical Experiment: Greenhouse gases in the atmosphere produced by modern man are changing the global radiation balance — Carbon Dioxide and Climate: A Scientific Assessment. Climate Research Board, National Academy of Sciences, 1979. Resulting climate trends, however, depend on complex interactions between the ocean, atmosphere, and biosphere. Is it possible to harness existing knowledge of ocean and atmospheric circulation to build models accurate enough to forecast the climate consequences of a probable buildup of greenhouse gases continuing well on into the 21st century? by Kirk Bryan r rofessor Roger R. Revelle, former director of the Scripps Institution of Oceanography and doyen of ocean scientists, calls the buildup of greenhouse gases* in the atmosphere "Man's greatest geophysical experiment." A greenhouse gas in the atmosphere absorbs the invisible, long-wave radiation to space that serves as the Earth's indispensable air-conditioning system. Hence, the greenhouse gases tend to trap heat within the atmosphere, and, in the absence of negative feedback, cause climate warming. The oceans are a very important component of the incredibly complex climate system. This is particularly true when climate is in the process of changing. Is it possible to synthesize existing knowledge of the oceans and atmosphere into mathematical models in order to provide a guide to future global changes in temperature and rainfall patterns? Providing such a forecast is the most intellectually challenging problem now confronting Earth scientists. The most ubiquitous natural greenhouse gas of all is the natural blanket of water vapor in the Earth's atmosphere. Almost everyone has experienced the great difference in night-time cooling between a dry desert climate and a moist humid climate. Revelle's attention was first attracted to this problem through his pioneering work on the global carbon cycle. Carbon dioxide is increasing, but it is still only a small constituent of the atmosphere. However, there is a limited band of wavelengths in which water vapor does not absorb outgoing long-wave radiation. This band from 7 to 20 microns, called the water vapor "window," is shown in Figure 1 . Outside of this window, the transmission of thermal radiation in the lower atmosphere is effectively zero. It is within the window that carbon dioxide plays its important role. Carbon dioxide has a high transmission over most wavelengths, but a very low transmission band within the water vapor window. The greenhouse Roger R. Revelle, director emeritus of Scripps Institution of Oceanography, was awarded the 1986 Balzan Prize for oceanography and climatology from the International Balzan Foundation of Milano, Italy. Revelle pioneered research in oceanography and atmospheric conditions and was one of the first scientists to recognize changes in the atmosphere due to the "greenhouse effect. " * Radiatively active gases in the atmosphere which are transparent to incoming solar radiation, but which absorb outgoing long-wave radiation, are called greenhouse gases. Carbon dioxide produced by the burning of fossil fuels is the most important man-made greenhouse gas. 36 Can We Model the Consequences? Wavelength /microns Figure 1. Wavelengths radiated by the Earth (dashed line) with percent absorption by atmospheric gases (solid line). The majority of radiation emitted from the Earth's surface lies in the wavelengths ranging from 4 to 80 microns, the infrared or heat portion of the electromagnetic spectrum. Although radiation of some wavelengths, notably those between 8 and 12 microns, is able to pass directly to space unless intercepted by cloud (clear area on graph), most of the outgoing radiation is absorbed by the atmosphere— principally by water vapor, and, in the "water vapor window," at least partially by carbon dioxide. (After K. L. Coulson, 1975. Solar and Terrestrial Radiation, Academic Press) effect of carbon dioxide is particularly important because it absorbs long-wave energy in this window, blocking the escape of long-wave energy to space, and making the Earth's natural air conditioning system less effective. Recently it has become evident that other greenhouse gases, such as the chlorofluorocarbons (freons to the layperson), are also rapidly increasing in the atmosphere. The effect of these different gases is thought to be additive. Thus, the combined contribution of these other gases taken together could soon be comparable to the greenhouse effect of carbon dioxide alone. Supercomputers and Modeling The climate response to greenhouse gases depends on the interplay of many processes. For example, any warming of the atmosphere allows an increase of water vapor that in turn increases the greenhouse effect. Likewise, the melting of snow and ice decreases the reflectivity of the earth's surface and enhances climate warming. Many of these effects can be studied in the context of models of the atmosphere alone. Extremely detailed mathematical models of the atmosphere have been developed and are now being used routinely in medium-range (4 to 10 days) forecasting by the U.S. Weather Bureau and the European Center for Medium-Range Weather Forecasting, located near Reading in England. Some components of these models, such as the radiation balance and the equations of motion for large-scale atmospheric flow, are developed from basic physical principles and can be verified in detail. Others, such as the formulation of clouds and precipitation, are more empirical, and are still in the process of development. A key factor in recent progress in atmospheric models has been the great increase in the power of supercomputers. One of the most detailed models of the atmosphere now in existence is being used at the European Center for Medium- Range Forecasting, where European countries have pooled resources and scientific manpower in a remarkably successful venture. As the power of computers keeps growing, it will become feasible to use these very detailed models as one of a variety of tools for testing climate sensitivity to greenhouse gases. The ocean plays two very important roles in a greenhouse gas-induced climate change. First, the ocean absorbs carbon dioxide. In fact, a 1983 National Academy of Science report estimated that about half of the carbon dioxide produced by fossil fuel burning has already been taken up by the ocean. The growing importance of other greenhouse gases, which are not absorbed to the same extent by the ocean, make the carbon dioxide uptake by the ocean less crucial than previously thought. The second role of the ocean, however, remains extremely important. This is related to the ocean's enormous ability to store heat. The thermal inertia of the ocean is very important in the transition from one climate regime to another. A complete adjustment of the ocean to a new climate requires thousands of years. Indeed, the deep ocean Is 37 The Little Ice Age ^^ur years are turned upside down; our summers are no summers, our harvests are no harvests," declared John King, an Elizabethan preacher in 1595. This bad weather was part of the Little Ice Age, when, although the land was not covered by ice sheets as in past geological ice ages, for a few hundred years the weather in Europe and America was considerably worse than today. Thus, although we tend to think of climate as stable, in fact, it can and does change — as the Little Ice Age attests to. The Little Ice Age began near the end of the European Medieval period. The earliest known advances of mountain glaciers that heralded the Little Ice Age date to the 12th and 13th centuries. An early phase of cold climate lasted until the end of the 15th century, when a minor climate reversal occurred. The main phase of the Little Ice Age came when, in the early 17th century, many glaciers reached their greatest extent since the last glacial age. For some glaciers, the maximum was reached even later, in the early 19th century. About 1850, a general warming trend began and caused glaciers to recede dramatically, with minor halts or reversals occurring in the 1880 and 1890s, and again in the 1 920s. During the cold centuries of the Little Ice Age, the sea ice spread southwards. The Ice edge at one time extended beyond Iceland to the Faeroes Islands, which lie 200 miles northwest of the Shetland Islands, and reportedly even allowed a polar bear to land there from the ice pack. As might be expected, ice played a large part in the change of climate during this period; but, whether this is cause or effect remains unclear until research into climate change reveals more about the mechanisms behind the large- and small-scale weather variations. This climate change not only varied considerably from decade- to decade, but also affected various parts of the world differently. 6°C Climatic Optimum 6 4 2 0 THOUSANDS OF YEARS BEFORE PRESENT Climate of the past 10,000 years. The general trends in global temperature have been estimated from geological records of mountain glaciers and fossil plants. During the Climatic Optimum, temperatures were about 2 degrees warmer than at present. About 300 years ago, during a climatic episode known as the Little Ice Age, temperatures were cooler than at present. (After Imbrie and Imbrie, 1979, Ice Ages, Enslow Publ.) Winters in China and japan were actually milder during the Little Ice Age, because weaker winds from the west allowed them to benefit more from the warm Pacific air. The American Midwest was cooler and wetter than before or after the Little Ice Age, because the wet Pacific winds traveled further south across the country. For the latter part of the Little Ice Age, deductions from old archives are paralleled by evidence from a worldwide network of meteorological observing stations, which have produced up to 300 years of continuous records of barometric pressure, wind speed, temperature, and rainfall. By reconstructing maps of typical air pressures over the North Atlantic and Europe, and then studying patterns of high and low pressure that produce these winds and weather, hlubert Lamb, chairman of probably still adjusting to the present climate from the colder climate that prevailed during "the Little Ice Age" in the 1 7th and 1 8th centuries. Unfortunately, we have little quantitative data to guide us on how the ocean responds to long time- scale changes in climate forcing. This lack of data makes it difficult to verify the behavior of ocean models designed to estimate the thermal inertia of the ocean. Many experts feel that the possible role of the ocean in delaying the onset of climate change is one of the most uncertain factors in the entire question of the effect of greenhouse gases on climate. Oceanographers have far less data for guidance in building models than meteorologists. Nevertheless, a remarkable increase in information has resulted from programs such as the Mid Ocean Dynamics Experiment (MODE), Ceochemical Ocean Sections Study (GEOSECS), and Transient Tracers in the Oceans (TTO). The new data provided by these 38 the World Meteorological Organization, has shown that during the winters around 1800 the winds coming into western Europe from the Atlantic were weaker than those of the first 40 years of the 20th century. A weaker flow of air from the ocean would have made western Europe more vulnerable, in winter, to cold air from Scandinavia and northern Russia. The lack of vigor in the warm Atlantic winds also may help explain the weaker behavior of the Gulf Stream and the extent of the sea ice that spread from the Arctic. Temperatures of the North and South Atlantic sea surface, as measured by British and American ships from 1780 onward, show that, during the Little Ice Age, the warm current of the Gulf Stream took a more southerly course across the Atlantic and swung further south than it does now as it approached the coast of Europe. It was as if the countries of western Europe had moved 300 miles to the north, or, more precisely, the pattern of global winds, including the jet stream, had shifted south. The weather pattern also shifted eastwards or westwards. Eor example, around the 1740s, Britain had relatively dry summers while Moscow was repeatedly wet; three decades later, the situation was reversed. The wavy motion of the jet stream, blowing from west to east over the stormy zone, but also zigzagging north to south, helps determine the wet and dry areas around the world. These areas are a compromise between the attempts of the jet stream to keep to a consistent wavy pattern right around the world, and the mountains, plains, oceans, and ice packs that impose their own pattern. During the Little Ice Age, four such zigzags were probably common, as opposed to this century's pattern of three, because the jet stream was weaker and traveling further south. If so, the area of warm southerly winds would have been pulled westwards into the Atlantic, while Europe suffered wintry blasts from the PRESENT LITTLE ICE AGE The southward shift of the warm Atlantic water during the Little Ice Age. The Gulf Stream, which sweeps across the Atlantic and delivers warm water to Europe, was less effective. Arctic. Such patterns of east-west variation, linked to the jet stream, also explain why the worst winters as recorded in Moscow and London seldom coincided. Although the Little Ice Age was 'little' in severity — yearly average temperatures for that period were 8.8 degrees Celsius as compared to 9.4 degrees Celsius for the first 50 years of this century — and relatively brief in duration, studying its climatic changes could help us forecast what may occur in the future, hlowever, this cooling trend with its enormous variety of weather, serves notice of the manifest complexities involved in predicting the next 300 years of climate. — Eleanore D. Scavotto programs are particularly important for showing us the similarities and differences between the dynamics of the ocean and the atmosphere. These insights are in turn important in deciding which features of atmospheric numerical models can be adapted for the ocean. The data base for modeling will be vastly improved with the measurements planned for the World Ocean Circulation Experiment (WOCE), an international program to take place from 1987-1995 (page 25). Coupled Models Climate variability on longer time scales can not be understood without taking into account the interaction of the ocean and atmosphere. This has motivated the building of a hierarchy of coupled models, ranging from the very simple, and somewhat abstract, to the increasingly detailed and complex. The more complex models of atmospheric and ocean circulation are similar to those used in numerical forecasting models, but as yet are far less 39 Figure 2. Effective modeling requires that complex systems be simplified. The numerical model of climate developed at the Geophysical Fluid Dynamics Laboratory, while it takes into account both the circulation of the ocean and atmosphere, is based on a highly idealized geography of the Earth. The atmospheric circulation is forced to be alike over each pair of ocean and land sectors, and the ocean circulation is alike in each ocean sector, v\/ith mirror symmetry across the equator. The atmospheric component includes individual cyclones and anticyclones, as we// as a specification of rain and snowfall, and a detailed treatment of radiation. The ocean component simulates the separate contributions of wind action and of heat and salinity sources to the ocean circulation. (After Spelman and Manabe, J. Geophys. Res. 89, 1984). detailed than the most advanced atmospheric models, such as those at the European Center. The simplified geometry of a coupled model developed vi'ith my colleague, Syukuro Manabe, at the National Oceanic and Atmospheric Administration's Geophysical Fluid Dynamics Laboratory in Princeton, New Jersey, is shown in Figure 2. Similar coupled models also have been developed at Oregon State University and the National Center for Atmospheric Research, Boulder, Colorado. In the idealized geography of the Princeton model, each hemisphere is divided in six sectors which are alternately land and ocean. It is assumed that the circulation has a mirror symmetry across the equator and that it is the same in each of the three 1 20- degree longitude sectors around the hemisphere. This simplification reduces the calculation to only '/& of that required for an entire globe. The atmospheric component of the coupled model predicts the flow of momentum, heat, and precipitated water into the ocean, as well as the loss of surface water as a result of evaporation. The atmospheric model has been used without coupling it to an ocean model by Manabe in many studies testing the sensitivity of climate to changes of carbon dioxide in the atmosphere. In coupling it to an ocean model, even the effect of river runoff on the salt balance of the ocean is taken into account in a simplified way. Where the model predicts temperatures below the freezing point of seawater, the numerical ocean model forms a thin layer of sea ice which tends to impede further heat exchange between the ocean and atmosphere. The numerical representation is somewhat coarse (400 by 400 kilometers). The atmospheric component resolves individual storms, but the ocean model cannot resolve individual eddies, like the meanders in the Gulf Stream. Nevertheless, it does provide a simple representation of the thermohaline (density-driven) and wind-driven components of the ocean circulation. Numerical Experiment A numerical experiment is shown in Figure 3. The ordinate is the average sea-surface temperature. The lower horizontal dashed line represents the average sea-surface temperature for an equilibrium climate with normal atmospheric carbon dioxide content. The upper horizontal dashed line represents the average sea-surface temperature for a model climate with four times greater carbon dioxide. The first stage of the calculation involves finding these two equilibrium climates by adjusting the ocean and atmospheric components of the model until a steady heat balance is achieved. It has been possible to find simulated climates in which the net drift of the ocean model was the equivalent of a flow of heat through the ocean surface of less than 1 watt per square meter. This is equivalent to a change of deep ocean temperature of 2 degrees centigrade per thousand years. The second stage of the calculation is represented by the solid line in Figure 3. Starting with the steady climate for normal carbon dioxide, a numerical integration of the coupled model simulates the climate response when the carbon dioxide is suddenly increased by a factor of 4. The inertia of the ocean's heat capacity prevents an abrupt warming after the sudden increase of Figure 3. A numerical experiment to test the response of the Earth's climate to a rapid rise of atmospheric carbon dioxide. The ordinate is the average sea-surface temperature of the model ocean. The abcissa is a measure of time after a sudden increase of atmospheric carbon dioxide. The main factor in slowing the response is the thermal inertia of the ocean. 40 (TJ E \ Hi cc C/) (/) LU 45 LATITUDE Figure 4. The zonally averaged excess temperature in the atmosphere and ocean in degrees Celsius predicted by the model one decade after the sudden increase of atmospheric carbon dioxide. Note the deep penetration of the excess heat into the ocean between 50 and 60 degrees North latitude. (From Spelman and Manabe, |. Ceophys. Res. 89, 1984) atmospheric carbon dioxide. The speed at which warming near the ocean surface takes place depends on the rate of penetration of excess heat into the ocean. At present, the vertical pathways by which surface influences find their way downward into the main thermocline* are a matter of intense interest to physical and chemical oceanographers. A picture of penetration in the model 10 years after the sudden carbon dioxide increase is shown in Figure 4. In the upper part of the panel, the temperature changes that have taken place in the atmospheric model can be seen. Note that the largest change takes place in the lower atmosphere in polar regions. This is associated with a warming of a very cold dome of air over the pole. This "polar amplification" takes place primarily in winter. Even though the model predicts rather spectacular temperature increases, the actual impact is uncertain, and may in fact be rather minimal. This is because, while the ice pack may be somewhat reduced, temperatures will remain quite cold, and the ocean effects may be little changed. The greater effects will likely be in the subarctic regions. * The layer containing the strong vertical temperature gradient. It is in subarctic latitudes where the stratification of the ocean is weakest, and penetration of excess heat down into the ocean is the greatest. In other areas, the penetration of excess heat is reduced by greater stratification of the water column. In the vicinity of the pole, for example, fresh water lowers the density of surface waters. In the tropics, very warm water at the surface also makes the water column stratified, and stable. The model's results support an earlier estimate (1979) by the National Academy of Sciences that the full impact of the present increase of greenhouse gases will be delayed several decades by the thermal inertia of the oceans. In the late 1950s and early 1960s bomb tests released a large amount of radioactive material in the atmosphere that eventually fell out on the ocean. The penetration of these radioactive, transient tracers appears to provide a source of empirical data with which we can judge the realism of the penetration of excess heat in our coupled climate model. It had been suggested that transient tracers would provide an overestimate of the penetration of excess heat because a warmer climate would make surface waters less dense and retard downward mixing. To check this idea, a passive tracer was introduced into the coupled model equivalent to the transient radioactive tracers caused by the bomb tests. The net penetration of this tracer into the 41 ocean model was thus compared to the penetration of the excess heat. In numerical experiments, like any other experiments, one has to be prepared for the unexpected! The thermal anomaly actually penetrated faster than a passive tracer introduced at the surface. The explanation turns out to be relatively simple. Climate warming in the model does not seriously modify the normal downward ventilation of heat into the main thermocline, but it does impede the upward pathways which normally bring heat to the surface in subarctic latitudes. Convection, which brings heat upward, is particularly sensitive to surface warming, and a weakening of convection and the thermohaline circulation therefore enhances the ability of the model ocean to absorb excess heat. "In the Hands of the Builders" The implications are interesting. If climate warming causes convection in subarctic latitudes to weaken and traps more heat below the surface than normal, the total uptake of heat by the ocean is enhanced. The delaying effect of the ocean will be greater. Consequently, the onset of a climate change in the atmosphere because of increasing carbon dioxide will be later than predicted. A healthy skepticism must always be maintained toward all idealizations of reality, and coupled ocean/atmosphere models are no exception. These climate models can best be thought of as a collection of models. A necessary condition for the quality of the overall model is that each component has been carefully designed and tested before inclusion. In general, atmospheric models like the one used in the coupled model at the Geophysical Fluid Dynamics Laboratory have been extensively tested against climate data for the atmosphere. Less verification and testing have been carried out for numerical models of ocean circulation. Lewis F. Richardson, a distinguished English mathematician and meteorologist, anticipated most of the elements of a coupled ocean-atmosphere climate model in a remarkable monograph (Weather Prediction by Numerical Process, 1922, Cambridge University Press, 236 pp.) written more than 60 years ago. In regard to model making in the absence of complete information, he writes of the necessity to ". . . .carry on business on premises which are, so to speak, in the hands of the builders." These words hold true today, and we must be prepared to revamp our models as new data become available. A very comprehensive test of the assumptions of current ocean models will be provided by the data sets to be collected by the World Ocean Circulation Experiment (WOCE). Kirk Bryan is an Oceanographer witti the Geophysical Fluid Dynamics Laboratory of the National Oceanic and Atmospheric Administration, Princeton, New Jersey. Special Student Rate! We remind you that students at all levels can enter or renew subscriptions at the rate of $17 for one year, a saving of $5. This special rate is available through application to: Oceanus, Woods Hole Oceanographic Institution, Woods Hole, Mass, 02543. Attention Teachers! We offer a 40-percent discount on bulk orders of five or more copies of each current issue — or only $3.00 a copy. The same discount applies to one-year subscriptions for class adoption ($14.20 per subscription). Teachers' orders should be sent to Oceanus magazine, Woods Hole Oceanographic Institution, Woods Hole, MA 02543. Please make checks payable to W. H.O.I. Foreign checks should be payable in dollars drawn on a U.S. bank. Selected References Bryan, K., F. G. Komro, S. Manabe, and M. ). Spelman. 1982. Transient climate response to increasing atmospheric carbon dioxide. Science 215: 56-58. Bryan, K., and M. J. Spelman. 1985. The ocean's response to a carbon dioxide-induced warming. /. Ceophys. Res. 90 (C6): 679- 688. Carbon Dioxide Assessment Committee, Board of Atmospheric Sciences and Climate. 1983. Changing Climate. 496 pp. Washington, D.C.: National Academy of Sciences. Schlesinger, M. E., W. L. Gates, and Y-.). Han. 1985. The role of the ocean in carbon-dioxide-induced climate change: Preliminary results from the OSU coupled atmosphere-ocean general circulation model. In Coupled Ocean-Atmosphere Models, ed. J. C. ). NIhoui, 767 pp. Amsterdam: Elsevier. Washington, W. M., and C. L. Parkinson. 1986. An Introduction to Three-Dimensional Climate Modeling. 422 pp. Mill Valley, California: University Science Books. 42 Orbital Geometry, CO2, and Pleistocene Climate The Earth wobbles, the sunlight changes, the ice sheets come and go. Scientists are beginning to understand the pattern. by Nicklas G. Pisias, and John Imbrie v^ne of the most dramatic features of the geological record of the last million years is the evidence of major advances and retreats of large continental ice sheets. For example, reconstructions of the Earth during the last ice age, approximately 1 8,000 years ago, show that all of Canada and parts of the northern United States were covered by a mass of ice that in places was more than 3,000 meters thick. The changes in global climate and the shifts in animal and plant communities that accompanied this increase in the volume of ice were substantial even at locations far removed from the ice itself. Many people find it surprising that the last ice The Earth's Orbit 23 V, March 21 June 21 December 21 September 23 Figure 1. The orbital geometry of the Earth. Viewed in the present, the tilted Earth (the axis of rotation is tilted 23'/2° away from a vertical perpendicular to the plane of orbit) revolves around the Sun along an elliptical path. Since the orientation of the axis remains fixed in space, changes in the distribution of sunlight cause the succession of the seasons, hlowever, the orbital geometry of the Earth is not fixed over time. The following three figures illustrate the basic components to variations in the orbital geometry, and consequently, to the amount of Sun's energy received at a given time (season) and place. 43 age was only one of a long series of climatic oscillations in which the Earth has gone back and forth between conditions similar to the last ice age and conditions similar to that of today. During the last million years, for example, the Earth has experienced about 10 major and 40 minor ice ages. The causes of such dramatic changes in the global environment have been the subject of much research and speculation during the last century and a half. Two major developments have occurred during the last 15 years. First, long-term (secular) variations in the Earth's orbit, we have learned, are the fundamental cause of the ice-age cycle. Second, changes in the atmospheric concentration of carbon dioxide, we know, are part of the complex of mechanisms that lead the Earth into and out of an ice age. In turn, these discoveries have raised questions about how the Earth's oceans and atmosphere interact to control our environment — not only on ice-age time scales, but on human time scales as well. Climate and the Earth's Orbit This December marks the 10th year since the paper "Variations in the Earth's Orbit: Pacemaker of the Ice Ages," was published in the journal Science by James D. Hays of the Lamont-Doherty Geological Observatory and his colleagues at Brown and Cambridge Universities. This paper presented strong statistical evidence supporting the hypothesis that variations in the pattern of incoming solar radiation were the fundamental cause of the succession of Pleistocene (ranging from 2 million to 18,000 years ago) ice ages, and thus the major mechanism for changing global climate on time scales between 10,000 and 100,000 years. Intuitively, the hypothesis that changes in the Earth's orbit control climate is not surprising. After all, the major climatic event experienced each year, the advance of the seasons, is caused by insolation changes related to the Earth's orbit (Figure 1). When the Earth's axis is tilted toward the Sun in one hemisphere, we have summer; when it is tilted away from the Sun, we have winter. Long-term changes in the orbit — changes that are caused by perturbations in the Earth's gravitational field as the planets change position — are somewhat more complicated than the annual revolution of the Earth around the Sun. Three distinct phenomena are involved: changes in the angle of (;7( of (he Earth's axis with respect to the orbital plane (Figure 2); changes in the eccentricity of the orbit (Figure 3, page 45); and changes in the season of the year when the Earth makes its closest approach to the Sun (perihelion). The last effect is the result of precession (Figure 4, page 46). Each of these components of orbital variation changes the incoming radiation in a particular way, and each component varies with a different and characteristic set of periods. The tilt of the Earth's axis varies between about 22 and 25 degrees, at periods close to 41,000 years. An increase in tilt concentrates the radiation toward the poles, and reduces the amount received each year at low latitudes. The climatic effect of precession is to change the distance between the Earth and the Sun at any AXIAL TILT a. Maximum Tilt Minimum Tilt DECEMBER 21 TILT = 23V2'' Figure 2. a) Variation in the tilt of the Earth's axis from maximum to minimum, b) The effect of axial tilt on the distribution of sunlight. When the tilt is decreased from its present value of 23V2°, the polar regions receive less sunlight, when the tilt is increased, polar regions receive more sunlight. (After Imbrie and Imbne, 1979) given season. Astronomers parameterize these changes in a precession index, which is in effect a measure of the Earth-Sun distance on June 21. These changes occur in two narrow frequency bands: one at periods near 19,000 years and the other near 23,000 years. At mid-latitudes, the effect of precession is to change the intensity of the incoming solar beam on the order of 10 percent above or below the mean value for a given season. The present orbital configuration is such that, during summer in the Northern FHemisphere, the Earth is farther away from the Sun than during winter. Less radiation therefore reaches the Earth's upper atmosphere in northern summers than in southern summers. As the Earth's axis precesses, the position of each season changes. About 12,000 years ago, for example, summertime radiation was greater in the Northern than in the Southern Hemisphere (Figure 5, page 47). The shape of the Earth's orbit varies from a nearly circular form (eccentricity equal to 0.00) to a considerably more elliptical form (eccentricity equal to 0.06). These changes occur in two broad frequency bands: one at periods near 100,000 years and one at periods near 400,000 years. Unlike the variations in tilt and precession, variations in orbital eccentricity change the annual total radiation 44 ECCENTRICITY Earth i S Changing Shape Of The Ellipse Ellipses With Different Eccentricities Figure 3. The orbit of the Earth changes shape from nearly circular to more elliptical. This is termed eccentricity, and expressed as a percentage. (After Imbrie and Imbrie, 1979) received at the top of the Earth's atmosphere — but only by a small amount, about 0.1 percent. The main effect of the eccentricity cycles is to modulate the amplitude of the precession cycles. When eccentricity is high, the effect of precession on the seasonal cycle is strong. When the orbit is circular, the position along the orbit at which the solstices or equinoxes occur is irrelevant. All three cycles — eccentricity, precession, and axial tilt — are compared in Figure 6, page 47. The idea that variations in the Earth's orbit control long-term climate change is not new. Scottish geologist James Croll first presented it in the 1860s. Croll argued that insolation changes in the winter were critical to the formation of ice sheets, and predicted that times of cooler winters marked times of ice accumulation. Stressing the effect of precession over tilt, he predicted that ice sheets would grow in one hemisphere while they decayed in the other, and that eccentricity would have a major influence on climate. The orbital hypothesis was first quantitatively formulated by the Serbian mathematician Milutin Milankovitch in the 1920s and 1930s. Milankovitch calculated how variations in the orbit affected the distribution of solar insolation received by the Earth. Unlike Croll, he argued that insolation changes in the high northern latitudes during the summer season were critical to the formation of continental ice sheets. During periods when insolation in the summer was reduced, the snows of the previous winter would tend to be preserved — a tendency that would be enhanced by the high albedo (reflectivity) of areas covered by snow and ice. Eventually, the effect of this positive feedback would lead to the formation of ice sheets. Because the orbit of the earth changes in a regular and predictable manner, a simple (linear) version of the Milankovitch hypothesis would predict that the amount of continental ice would vary with the same regular pattern. Such a model would predict that the record of continental ice variations would contain those frequencies — and only those frequencies — of orbital components that are responsible for changing the distribution of incoming solar radiation, that is, the 41,000-year period of tilt and the 19,000- and 23,000-year periods of precession. No 100,000-year or 400,000- year climatic cycles related to eccentricity would be expected, because changes in this orbital parameter affect not the frequency but the amplitude of precession variations. Investigations during the last 1 5 years have demonstrated two things. First, the 19,000-year, 23,000-year, and 41,000-year periodicities predicted by the Milankovitch hypothesis actually occur in long records of Pleistocene climate. Second — and rather surprisingly — they demonstrate that the climatic oscillations observed in these frequency bands are linearly related to (that is, highly coherent with) the orbital forcing functions (Figure 7, page 48). These findings would certainly have pleased Croll and Milankovitch, the fathers of the orbital theory of climate. Although much work remains to be done on the mechanisms by which insolation changes driven by orbital variations actually produce the succession of Pleistocene ice ages, there is no doubt about what the fundamental cause of this succession is. The 100,000-Year Climate Cycle So far, so good. But there are no grounds for complacency. The same investigations that identified climatic cycles related to precession (19,000 and 23,000 years) and tilt (41,000 years) also have demonstrated that the largest climatic cycle during the last million years is linearly (and therefore presumably causally) related to the 100,000-year cycle of eccentricity. As previously explained, this cycle was not predicted by Milankovitch and, in fact, cannot be related to the orbital forcing by any simple, linear mechanism. Although Milankovitch would have been surprised by this fact, it is interesting that Croll — the semi-quantitative geologist from Scotland — concluded in 1864 that the major features of Pleistocene climate could be explained as an effect of eccentricity cycles. Although the papers of Croll 45 AXIAL PRECESSION OR 'WOBBLE' Axis of Rotation Orbital Plane PRECESSION OF THE ELLIPSE Earth PRECESSION OF THE EQUINOXES Mar. 20 June 21 /^^ Sept. 22 Dec. 21 TODAY fAat.20 /"^^ "]'") Sept. 22 5,500 YEARS AGO June 21 Sept. 22 Dec. 21 11,000 June 21 YEARS AGO Mar. 20 • EARTH on December 21 ® SUN Figure 4. Precession has two comporients: a) Axial Precession — the Earth's axis of rotation "wobbles," like that of a spinning top, so that the North Pole describes a circle in space; and b) Elliptical Precession — (he elliptical orbit itself is rotating, independently and slowly. Taken together, the result is c) Precession of the Equinoxes — the positions of the equinox (March 20 and September 22) and solstice (June 2 1 and December 21) shift slowly around the Earth's elliptical orbit, completing one full cycle every 22,000 years. Today, the winter solstice occurs near the perihelion (point on the orbital path closest to the Sun). Eleven thousand years ago, the winter solstice occurred near the opposite end of the orbit. (After Imbrie and Imbrie, 1979) are now rarely quoted, the problem first posed by him — the identity of the mechanisms by which the Earth's climate system responds so dramatically to variations in eccentricity — remains a major scientific issue. Although the problem remains unsolved, recent evidence and arguments point in the direction of a solution to the mystery of the 1 00,000- year cycle of ice ages. Climate and Atmospheric CO2 As previously indicated, the 100,000-year climate cycle cannot be a linear response to orbitally-driven variations in insolation. But the cycle can be modeled as a nonlinear response. Perhaps half a dozen models of this type have already been suggested. Several are based on nonlinear processes in the accumulation and wasting of glacial ice — essentially on the fact that it takes longer for an ice sheet to grow than to decay. As an ice sheet melts, for example, the ocean level rises so that glaciers grounded along continental margins become floating ice shelves which easily break off and enter the ocean as icebergs. Another model is based on the nonlinear response of the Earth's crust to ice-sheet loading. As a glacier grows, the rock on which it sits subsides under the weight of the ice. Conversely, as an ice sheet melts the ground will rise. The rate at which the Earth's crust responds to the weight of an ice sheet is controlled by the size of the glacier and by the properties of material deep in the earth. Models proposed by W.R. Peltier of the Department of Physics at the University of Toronto, which include these processes, predict a glacial record which contains not only the periods of precession and tilt but also the 100,000-year period of eccentricity. More recently, another component of the climate system — the global carbon cycle — has been 46 implicated as being part of the complex of mechanisms by which the system responds to Milankovitch forcing, including its response to the 100,000-year cycle of eccentricity. Speculation that changes in carbon dioxide (and other "greenhouse" gases) are part of the causes of climate change (see pages 9 and 16) is not new. When the nature of long-term climate change became evident many decades ago, such mechanisms as changing solar activity and changes in atmospheric carbon dioxide were suggested as possible causes. However, such speculations were unsupported by data. Not until 1980, when the first measurements of carbon dioxide from ice cores were made (see page 55), was there direct evidence that there have been major changes in carbon dioxide levels. Measurements of the COj found in gas bubbles trapped in Greenland and Antarctic ice cores show that the levels of CO2 were on the order of 80 parts per million less during the last ice age than they are today. Wallace Broecker of the Lamont-Doherty Geological Observatory at Columbia University was one of the first to suggest that the rise in CO^ at the end of the last glaciation 18,000 years ago would have warmed the atmosphere and thus helped to melt the ice sheets. Such a positive feedback in the climate system would help account for the rapid melting of ice observed in the geologic record. How can this idea be tested? The 40 percent change in atmospheric CO2 levels during the last 18,000 years raises two questions: 1) what processes caused this change, and 2) how is this change related to longer-term variations in orbital geometry and climate? It can be argued that since significantly more carbon dioxide is dissolved in the surface ocean than is found in the atmosphere, it is the level of COj in the surface ocean that ultimately controls the levels found in the atmosphere. Thus, the answer to the first question must lie in the ocean. Shortly after the ice-core data on carbon dioxide appeared, Broecker argued that changes in the nutrient levels of the ocean could account for the observed changes in CO2. If nutrient concentrations increased in the surface ocean, then biological activity and the fixing of carbon dioxide into organic matter would increase. As this fixed organic carbon was removed from the surface ocean to the deep ocean by settling, the level of COj in the surface ocean would decrease. As the atmosphere is in equilibrium with the ocean, the atmospheric COj would also decrease. However, geochemical data from deep-sea sediments do not support the hypothesis that the glacial ocean contained more nutrients than the present ocean. Other models to explain the observed CO2 records do not involve changing the amount of nutrients in the ocean. Instead, they postulate changes in the efficiency with which organisms utilize nutrients. Fanny Knox and Michael McElroy of Harvard University suggest that changes in the available sunlight in high-latitude regions of high productivity — changes which can be caused by variations in the Earth's orbital parameters — could increase the efficiency with which nutrients are fixed in the higher latitudes of the ocean compared with other latitudes, and thus change the rate at which .^ bO 5 5 40 >. -1 30 - -> 20 < Z 0 10- -J Z 0 -8 ■' Figure 5. Changes in solar insolation over time and latitude are related to variations in the Earth's orbit. Here, an insolation anomaly (a difference between past insolation and modern insolation) for the month of July was calculated over the past 30,000 years for latitudes from 60°S to 60°N. Several features are evident: a) An insolation peak occurs in the Northern Hemisphere approximately 1 1,000 years before present — the end of the last ice age, b) There is a minimum insolation level at 18,000 years before present — during the last ice age, and c) The insolation anomalies are strongly related to latitude. (After W. L. Prell, Geophysical Monog. 29) carbon dioxide would be pumped to the deeper ocean. The difficulty with this model and other models so far proposed is that they all predict that the deep ocean would become very depleted in oxygen — a prediction not supported by any geological data. All models that describe possible mechanisms for explaining the CO2 data predict changes in the distribution of the stable isotopes of carbon-1 2 and carbon-13. They predict that the difference between 250 T 200 150 100 50 0 -50 100 HOUSANDS OF YEARS BEFORE PRESENT Figure 6. Variations in orbital geometry as a function of time, according to calculations by A. Berger. Two components are dominant: precession of the equinoxes, and the axial tilt. Eccentricity, a secondary factor, mainly affects the amplitude of the precession cycle, and is included as a term in calculating the precession index, where minus numbers indicate a closer distance. Although these orbital variations have a small effect on the total energy received by the Earth each year, they have a larger effect on changing the seasonal and latitudinal distribution of that energy. 47 a O 300-1 o O < 1 0 25 50 75 100 125 150 Years BP x 1000 Figure 7. Changes in atmospheric CO2 and the gradient of carbon isotopes in the ocean as measured in planktonic and benthic foraminifera. Shaded area represents the field of data for atmospheric carbon dioxide as measured in ice cores (from A. Neftel and others. Nature 295; 220, 1982). Solid line is record of the difference in carbon isotope ratios measured in benthic and planktonic foraminifera in an eastern equatorial Pacific core. (From Shackleton and Pisias, "The Carbon Cycle and Atmospheric CO2 Natural Variations Archean to Present," Geophysical Monograph 32, American Geophysical Union, 1985) the isotopic ratios in nutrient-depleted surface waters and waters of the deep ocean should increase as the level of atmospheric CO2 decreases. Nicholas Shackleton and his colleagues at Cambridge University in England tested this idea by measuring the carbon isotopic ratio in benthic and planktonic foraminifera which were obtained from the same set of samples taken from a core located in the eastern equatorial Pacific. They postulated that the foraminifera which live in the near-surface regions of the ocean deposit their calcium carbonate shells in equilibrium with seawater, whereas foraminifera living at the seafloor preserve a record of the isotopic composition of the deep water. The calculated benthic-planktic difference is therefore a measure of the vertical gradient in carbon-12 and carbon-13 — which in turn may be related to the fraction of upwelled carbon removed by photosynthesis, and the amount of CO2 removed from the surface ocean and the atmosphere. Comparison of this isotopic gradient with ice core records of carbon dioxide suggest that the measured carbon gradient has the potential for estimating past changes in atmospheric chemistry. Because it is much easier to get long geological records from the ocean floor than it is from polar ice caps, this idea has prompted widespread interest in obtaining and studying long sedimentary records of carbon-isotope gradients. The longest record studied so far spans the last 350,000 years. Records of this length allow us to examine the second question posed earlier, namely, how do changes in the carbon cycle relate to orbital and climatic change? Three important results are suggested by early investigations. First, like the record of ice volume changes, the record of long- term CO2 change (or at least the carbon-gradient estimate of this quantity) contains periodic components linearly related to variations in eccentricity, tilt, and precession. Second, changes in the carbon budget lag changes in orbital geometry and lead changes in ice volume. Thus, changes in atmospheric CO2 are now seen as part of the complex of mechanisms by which changes in the Earth's orbit cause major changes in global climate. This hypothesis can be tested using a simple model describing ice volume changes as a function of changes in July insolation at 65 degrees North. The model assumes that the rate of glacial growth is slower than the rate of decay. In Figure 8, page 49, we show the time series of global ice volume predicted by this model, compared to the ice volume record from deep-sea sediments. The two series are generally similar. FHowever, there are some important differences. For example, the dominant mode of variation in the model is associated with the 41 , 000-year period of tilt; and the levels of some of the minima in ice volume are not well simulated. The figure also shows the predicted record of ice volume if we include CO2 change as part of the model. FHere, the CO2 record is added to the insolation record, and the model run using the same parameters as before. The fit of the model to the observed data is significantly improved, thus supporting the idea that changes in atmospheric CO2 play a role in causing climate change. Conclusions The geological data obtained from deep-sea sediment cores provide compelling evidence that variations in the Earth's orbit played a dominant role in changing climate over the last million years. There also is growing evidence that changes in the carbon budget, and the associated changes in atmospheric carbon dioxide levels, played a significant role in controlling Pleistocene climate. Why the 100,000- year cycle of eccentricity played such a dominant role in changing Pleistocene climate is a major scientific problem. Although the addition of CO2 variations to a simple climate model greatly improves the ability of the model to describe the geological data, we are still faced with the origin of the long period components observed in the CO2 time series. To the extent that we have answered part of the question of why ice sheets come and go with a period of around 100,000 years, we have raised the same question about the causes of changes in the global cycle of carbon. The idea that the 100,000-year cycle has its origin in the response of the Earth's crust to ice-sheet loading is a possibility. However, this hypothesis runs into problems if we consider the variations in the Earth's climate over time intervals of several million years. Data from cores taken as part of the Ocean Drilling Program (see Oceanus Vol. 29, No. 3, page 36) show that the time series of climate change during the period of 1 million to 2 million years ago do not contain a significant amount of variability associated with the 100,000-year cycle of eccentricity. (Indeed, if Milankovitch had lived a million years ago, he would have been right when he predicted that long-term climate change would be dominated by variations in orbital tilt.) Thus, any model that aims to describe major changes in climate as a function of orbital forcing, must not only account for the last million years of Earth history, but also must address the problem of the long-term 48 Orbits Only O -T 4- 5- Figure 8. Orbital geometry alone does not explain climate change. While modeling of global ice volume (top graph) is similar to the ice-volume record from deep-sea sediments (middle graph), the (it of the model to the observed data is improved when atmospheric CO2 concentration is added to the model (bottom graph). 100 150 200 250 3OO Age X 1000 Years evolution in the system's response to Milankovitch forcing. Undoubtedly, the next decade of observation and modeling will yield deeper insights into the mechanisms of climate change on all time scales, from a century to 100,000 years and beyond. Because the fundamental cause of the Pleistocene ice ages is known, investigators working on geological time scales have a major opportunity to establish how sensitive the Earth's climate system is to known changes in the radiation budget. They also can hope to discover some of the mechanisms by which Nature altered the concentrations of carbon dioxide in the Earth's atmosphere long before the Industrial Revolution. Nicklas C. Pisias is Associate Professor of Oceanography at the College of Oceanography, Oregon State University, lohn Imbrie is Doherty Professor of Oceanography in the Department of Geological Sciences, Brown University, Providence, Rhode Island. Selected References Imbrie, John. 1978. Geological perspectives on our changing climate. Oceanus 21(4): 65-70. Imbrie, ]., and K. P. Imbrie. 1979. Ice Ages — Solving The Mystery. 224 pp. Hillside, New Jersey: Enslow Publ. Imbrie, )., and ). Z. Imbrie. 1980. Modelling the climatic response to orbital variations. Science, 207:943-954. Imbrie, J., ). D. Hays, D. C. Martinson, A. Mclntyre, A. C. Mix, ). ). Morley, N, C. Pisias, W. Prell, and N. ). Shackleton. 1984. The orbital theory of Pleistocene climate: support from a revised chronology of the marine 0-18 record. In Milankovich and Climate, edited by A. Berger, ). Imbrie, I. Hays, G. Kukia and B. Saltzman. pp. 269-305, Hingham, Mass.: D. Riedel. Pisias, N. C., and N. ). Shackleton. 1984. Modelling the global climate response to orbital forcing and atmospheric carbon dioxide changes. Nature, 310:757-759. Shackleton, N. J., and N. G. Pisias. 1985. Atmospheric carbon dioxide, orbital forcing, and climate. American Geophysical Union, Geophysical Monograph 32:303-317. 49 The Polar Ice Sheets: A Wild Card in The Deck? by Stanley S. Jacobs It's getting hotter — or so most scientists predict — and you won't be able to go to the beach to cool off, because as the Earth warms, polar ice will melt and the surface of the ocean will expand, causing a sea- level rise of some 4.5 feet by the year 2030." That quote, from today's stack of mail, is typical of statements that we frequently encounter in the news media. A projected sea-level rise of 1 foot every 10 years can galvanize the attention of even a mountain dweller, but it is not a very likely scenario. Could the polar ice sheets really generate such a modern flood, given that they survived the last major deglaciation* largely intact, and have experienced only minor changes over the last several millenia? Probably not, but they should bear some close attention. Recent Sea-Level Rise Careful studies of regionally variable sea level records taken over the last century indicate a global rise of about a half inch per decade. The 5-inch (12- centimeter) global sea-level rise since the late 1800s is not tied to any melting of the ice caps, but can be * The last deglaciation began about 18,000 years ago, removed most of the ice on the northern continents, and was essentially completed 8,000 years ago. Above, the edge of the Ross Ice Shelf, standing 25 to 30 meters above the sea surface. The dye mark on the ice sheet provides a temporary navigational reference point for repeated oceanographic observations. 50 The Antarctic ice sheet is larger in area than the United States. The lightly shaded areas are the floating ice shelves. explained by thermal expansion of the ocean and the retreat of temperate mountain glaciers (Figure 1, page 52). The polar ice sheets thus do not appear to be the perpetrators of recent sea-level rise, this agrees with glaciological data, which suggest that Antarctica is slightly on the positive side of mass ■^v ^'X^fr m ^' Imfi 1^ H ^m iw" W^^^^B ;fl[ Ik ■c ^- ' ' ^ :.'*^^ ^Hi P ^K. '^^^H^B n /"' ' •■* t, ■ ^'•■■ •\ - ■ "^•*^?»k «. '-ylja •! ^/ \ An example of icebergs calving from a West Antarctic ice shelf. The largest iceberg is 32 kilometers in length. This enhanced Landsat image was taken several years ago, and shows rises, crevassed areas, and flow features. Periodic imaging of the same area would reveal changes in any of these features. (Courtesy of B. Lucchitta, U.S. Geological Sun/ey) balance. In other words, the precipitation being added to Antarctica is slightly more than the water and ice being lost — on a time scale of years or decades. In addition, analysis of crustal rebound and wander of the Earth's pole of rotation does not reveal a mass transfer away from the polar regions, as might be expected if the ice sheets were melting. Atmospheric Warming Global climate models project that atmospheric warming, as a result of increasing carbon dioxide and other "greenhouse" gases, will be greatest in the polar regions. The amount of warming, estimated to be several degrees Celsius (unless balanced by feedbacks not incorporated in the models), could reduce the area and thickness of sea ice. This would not directly influence sea level, but the now permanent Arctic pack ice might not reform if removed, as surface albedo (reflectivity) would decrease in concert with increased ocean heat flux and wind mixing of the surface layer. A warmer atmosphere will lead to a warmer ocean, after some lag, and both will eventually have some impact on the polar ice sheets. At this point, it is perhaps worthwhile to offer some orientation. We speak of "ice sheets" in the general s'^nse. In truth, there exists an important distinction. There is floating ice and continental ice. 51 80- ^ 60 E E Total Sea Level Sea Level Less Thermal Expansion Figure 1 . The role of thermal expansion in sea-level rise. Global sea-level change due to the melting of small glaciers is shown by the heavy solid line. Above it, the light line, is an estimated sea- level curve for increased atmospheric carbon dioxide, and subsequent rise in global ocean temperature. Subtracting the portion due to thermal expansion yields the dotted line curve. Thus, under these models, thermal expansion due to the warming of the oceans accounts for a substantial portion of the estimated sea-level rise. (After M. F. Meier, 1 984) The floating ice — the pack ice or sea ice of the Arctic, and the ice shelves of the Antarctic — represents large volumes of water, but since it is already floating, any melting will not change the sea level. A second distinction is worth noting here. The pack ice, or sea ice, of the Arctic is from the freezing of seawater, and is not considered under the term ice sheet. The ice shelves, on the other hand, are floating extensions of the continental ice sheets, and they form where ice flows into the sea. The continental ice, ice based on land, once melted and drained (or calved and melted) into the ocean, may, on the other hand, cause a sea-level change. The continental ice is distributed as follows: Antarctica, 91 percent; Greenland, 8 percent; and mountain glaciers (for example, in Alaska and the Alps), 1 percent. The present view is that while a global warming may increase melting in Greenland, the quantity of water produced will have only a minor impact on sea level. Therefore, we turn our attention primarily to the Antarctic. The West Antarctic Problem If Antarctica's mantle of ice could be instantaneously removed, one would find in the eastern longitudes a rugged continent mostly above sea level (similar to Greenland) and an archipelago with deep submarine basins and continental shelves in the western longitudes (Figure 2). This West Antarctic ice sheet, equivalent to a global sea-level increase of 5 to 6 meters if melted, is grounded on the islands and basins and nearly surrounded by floating ice shelves. The major portion of the ice sheet, including large parts of East Antarctica, drains through the ice shelves, which also retard the seaward progress of the outflowing ice streams. Thinning or removing these Texas-sized ice shelves would not change sea level, any more than a melting ice cube raises the level of the liquid in your cocktail glass. It could upset the dynamic regime of the ice sheet, however. ^ Larsen Ice Shelf 0 1 \ WEDDELL ,•'/ ) SEA AI .-Ronne Ice Shelf ^^ -Filchner Ice Shelf \ ^)^ George VI Ice Shelf ^^ KD ^ Ross Ice Shelf 1 ■ r 0 '000 \l^) km a 180 1 b ^x/ /V --2 ■-3 ^ \ 0 1000 2000 3000 . ," ' i tE) Figure 2. Antarctica, showing ice shelves (stippled) and areas where bedrock is more than 500 meters below sea level (hachured). A section along the bold line in 2a is shown in 2b, with the Ross Ice Shelf at left near the 180° meridian. The Filchner Ice Shelf is centered at 40° West. The Larsen and George VI Ice Shelves lie on the eastern and western sides of the Antarctic Peninsula, which is at 65° West. The ice is not static, but moving — at a few hundred meters per year — from higher elevations downslope to the ocean. (After Thomas and others, 1985.) 52 Figure 3. North-south vertical sections of salinity and temperature through the primary "warm" intrusion into the Ross Sea and beneath the Ross Ice Shelf. The lower diagram illustrates how the ice shelf may have been modified as it moves northward by surface accumulation and by basal melting and freezing. (In Oceanology of the Antarctic Continental Shelf, 1985) causing grounding lines* to retreat into the submarine basins and allowing ice streams to surge into the sea. There is evidence for a 6-meter higher stand of sea level 1 25,000 years ago. On millenial time scales, large changes in sea level are believed to exert a major control over the Antarctic ice sheet configuration and dynamics. However, there is a general concensus that the inverse problem, polar meltwater raising sea level, would take several hundred years to achieve a major impact. Stability of the Ice Shelves If the West Antarctic ice sheet is sensitive to the stability of its ice shelves, how sensitive are the ice shelves to climatic warming? Summer surface melting is now inconsequential along most of the coastline, except for the western Antarctic Peninsula, but several degrees of atmospheric warming would increase the runoff. Surface meltwater generated during warm periods on the flat ice shelves would be likely to percolate into the snow and refreeze. Ice is a good insulator, so heat conduction from the atmosphere should be slow and have minor effects on glacier flow. This allows us to dispel one of the common misconceptions about the melting and thinning of ice sheets. Indeed, the primary melting of floating ice comes from underneath rather than on top. The potentially vulnerable parts of the ice shelves are their vast basal areas, which are directly accessible to ocean heat flux. You may well ask why there would be any melting, since ice forms at the sea surface near ice shelves for much of the year, causing very cold (-1.9 degree Celsius), brine-enriched seawater to sink to great depths, where it is still found in summer. The primary reason is that the freezing point of seawater (T,,) decreases with increasing pressure, so that water overturned at the sea surface * Grounding lines mark the boundary between ice that is floating and ice that is grounded on the iand or seafloor (see Figure 2). Removal of the buttressing ice shelves might allow ice streams that now feed the ice shelves to flow much faster, or "surge." will be around 0.5 degrees Celsius above T,, at 700 meters depth. In addition to the melting potential afforded by this sensible heat bonus, the Tp depression may allow the redistribution of basal ice, with melting at depth followed by upwelling and deposition of ice at shallower levels. Another mass balance factor is the circumpolar deep water, which has access to the ice shelf bases at various locations around the continent. In the Weddell Sea and Ross Sea, its access is now limited to cooled, mid-depth intrusions across the continental shelf (Figure 3). Adjacent to the southwest Antarctic Peninsula, however, nearly unaltered, +1 .0 degree Celsius deep water penetrates beneath George VI Ice Shelf resulting in basal melting rates around 2 meters per year. A third reason to anticipate melting is that there is now known to be an active ocean circulation (and a functioning ecosystem) far beneath the largest ice shelves. The tidal and thermohaline circulations may regionally break down the stable stratification that would result from a layer of meltwater beneath the ice. Deep temperature and current measurements along the northern front of the Ross Ice Shelf have disclosed large seasonal and interannual variability. During winter, temperatures are expectedly lower but current velocities are markedly higher. In a southward-flowing "warm" current, monthly heat flux was highest during July (midwinter). Topographic steering and isopycnal* access to the deep water appear to control the source and recirculation of this water in the Ross Sea. Annual heat flux into the sub-ice shelf cavity appears capable of melting a few tens of centimeters, averaged over the ice shelf base. This is consistent with glaciological models of the steady-state ice shelf flow and mass balance. Beneath the ice, melting probably varies from a few meters per year near the ice front to several centimeters per year near the grounding lines, but there are large sectors where basal freezing prevails. * An isopycnal is a surface of constant density. 53 The Mass Budget of Antarctica On the attrition side of Antarctica's mass budget, estimates of ice shelf melting range from 0 to 675 cubic kilometers per year, while iceberg calving has been said to lie between 500 and 2,400 cubic kilometers per year. Most calving estimates derive from censuses of iceberg populations, an art best practiced by the Soviets and Norwegians. After accounting for duplicate observations, iceberg life expectancies (several years), the difficulty of measuring dimensions, and the probable errors in extrapolating from subsamples to the entire Southern Ocean, one is left with the major problem of temporal variability. The white continent simply does not calve icebergs as regularly as a white leghorn lays eggs. For example, the western third of the Ross Ice Shelf front has apparently not experienced significant calving since before Amundsen and Scott trekked to the South Pole in 1911. Ice shelf fronts in the Weddell Sea also had undergone decades of slow advance, until several large icebergs calved from the Larsen and Filchner Ice Shelves this year. The evidence suggests that the Weddell Sea breakout of 1986 does not represent a crisis event, such as the impending breakup of the ice sheets. Rather, indications are that what happened in the Weddell Sea this year has simply redressed a growing imbalance between advance and decay. The ice shelves of the region had been advancing slowly for decades. Accounts of the iceberg sizes differ, but they may contain substantially more water than the annual accumulation on the Antarctic continent. On the accumulation side of the Antarctic mass budget, precipitation minus evaporation and snow blown off the ice is around 2,200 cubic kilometers per year (± 10 percent). A warmer atmosphere can transport more moisture, and global climate models tend to show a few tenths of a millimeter per day more rapid increase in precipitation than evaporation at high southern latitudes. That does not sound like much, but Antarctica is a desert in terms of its "rainfall," which averages around 16 centimeters per year. With other factors unchanged, an increase of 0.2 millimeters per day would correspond to 1,000 cubic kilometers per year, half the present accumulation rate and twice the present rate of sea-level rise. There is evidence for both warming and increased precipitation on Antarctica in recent decades, but the records are still too short to separate climatic trends from decadal cycles. In terms of the ice sheet/sea level relationship, the potential importance of an increase in precipitation is that it could postpone the day of reckoning if not reverse the sign of the change. In the interim, the oceans might have time to absorb more of the greenhouse gases, and man might be able to reduce his pollution of the atmosphere. How to Play the Game Current estimates of iceberg calving, basal melting, and freshwater runoff (in the Greenland case) inspire little confidence that we really know how close the polar ice sheets are at present to a state of mass balance. However, radar and laser altimeters aboard polar-orbiting satellites could help considerably to narrow the range of uncertainties. Changes in the ice sheet margins and other features can be monitored with altimeter data in conjunction with radio-echo sounding profiles (for thickness) and navigational transponders (for velocity). Further, the altimeters should be able to detect relevant elevation changes on the continent, since a 1 centimeter change of global sea level would roughly correspond to a 2 meter elevation change over the grounded portion of West Antarctica. Large changes in ice shelf melting rates may be detectable in repeated thickness (elevation) profiles, but we also need to understand why there appear to be order-of-magnitude differences in melt rates between ice shelves. If a warmer atmosphere causes less sea ice formation on the continental shelves and less-dense shelf water, will that significantly change the shelf circulation or exchange with the deep ocean? Will global climate models with high latitude deep ocean convection, realistic sea ice cycles, and better-parameterized boundary layer processes show the same polar-amplified effects? While ocean models are formulated and then tested by relevant long-term observations, particularly of salinity, it is worth reflecting on the risky geophysical game we are playing with the Earth's climate. Antarctica may be a wild card in this game, but who can say the deck is not stacked — with Nature setting up the sting? Stanley S. Jacobs is a Senior Staff Associate at the Lamont- Doherty Geological Observatory of Columbia University, Palisades, New York. Selected References Giovinetto, M. B., and C. R. Bentley. 1985. Surface balance in ice drainage systems of Antarctica, An(arc(/c lournai of the United States 20{b):6-\ 3. Jacobs, S., ed. 1985. Oceanology of the Antarctic Continental Shelf, Antarctic Research Series No. 43. 312 pp. Washington, D.C.: Amer. Ceophys. Union. Meier, M. F. 1984. Contribution of small glaciers to global sea level. 5aence 226:1419-1421. National Research Council. 1985. Ad Hoc Committee on the Relationship between Land Ice and Sea Level, Committee on Claciology, Polar Research Board. Glaciers, Ice Sheets, and Sea Level: Effect of a COi-lnduced Climatic Change, Report of a Workshop, Seattle, Sept. 13-15, 1984. 330 pp. Office of Energy Research, Dept. No. DOE/EV/60235-1, Washington, D.C.: U.S. Dept. of Energy. Pourchet, M., F. Pinglot, and C. Lorius. 1983. Some meterological applications of radioactive fallout measurements in Antarctic snows. /. Ceophys. Research 88:6013-6020. Schwerdtfeger, W. 1984. Weather and Climate of the Antarctic, Developments in Atmospheric Science No. 15. 261pp. New York: Elsevier. Thomas, R. H., R. A. Bindschadler, R. L. Cameron, F. D. Carsey, B. Holt, T. J. Hughes, C. W. M. Swithinbank, I. M. Whillans, and H. |. Zwally. 1985. Satellite Remote Sensing for Ice Sheet Research, NASA Tech. Memo. No. 86233. 32pp. Washington, D.C.: NASA. 54 Polar Ice Cores Shallow ice core drilling atop Mt. Bird, Ross Island, Antarctica. by Julie M. Palais Locked in the Earth's polar ice sheets is a record of past climate stretching back at least 150,000 years. Studies of ice cores are providing scientists in many fields with a rich archive of information on past temperatures, precipitation, and atmospheric composition and circulation. The annual accumulation of precipitation on the ice sheets is preserved with little or no melting, except in coastal regions, and is gradually compacted into solid ice by processes of densification and recrystallization. Ice cores can therefore provide continuous atmospheric and climatic records, reaching back several hundreds of thousands of years. Glaciologists and paleoclimatologists who study ice cores do so in the hope that ". . . studies of the past may hold the key to the future . . . ." Studies of the climatic response to natural variations in carbon dioxide and atmospheric dust (continentally derived or volcanic), for example, may help scientists understand the effects of the buildup of carbon dioxide and other gases in the atmosphere as the result of man's activities. The longest and most complete ice core records have been obtained from the central regions of the two great ice sheets in Antarctica and Greenland (Figures 1, page 57, and 2, page 58). However, studies of cores from other parts of the Arctic (for example, Baffin Island, Alaska, and Spitsbergen) and high-altitude sites in temperate and equatorial regions (for example, the Quelccaya Ice Cap, Peru; Mt. Kenya, East Africa; and, most recently, the Tian Shan Mountains of north central China) have increased the geographical coverage and have provided important interhemispheric coupling for data interpretation. Ice cores preserve information on atmospheric composition and climate in three distinct chemical forms: 1) the stable isotope composition of the ice itself, determined primarily by climatic factors; 2) the soluble (for example, sulfate, nitrate, sodium, and chloride) and insoluble (mainly silicate particulates) impurities and heavy metals, such as lead, cadmium, and zinc, derived from marine, continental, biogenic, volcanic, and extraterrestrial sources; and 3) bubbles in the ice containing samples of the atmosphere at the time that the ice was formed. Studies of these parameters on both long (hundreds of thousands of years) and short (hundreds of years) time scales allow information to be obtained on large-scale climatic 55 The WO-foot-high Perito Moreno Clacier on Lake Argentina in the Glacier National Park, Argentina. Studies of ice cores from glaciers provide data on physical properties of glacier ice necessary for modeling flow characteristics. (Photo from UPl/Bettmann Newsphotos) fluctuations as well as on recent global pollution. The Ice Core Record Depth-Age Relationships. For the paleoclimatic and paleoatmospheric record to be of value, the age of the ice at each depth must be known. Dating methods that have been used include those based on ice flow models, radioisotopes, volcanic events, stable isotopes, and microparticles. In regions of relatively high annual accumulation (greater than 25 grams per square centimeter per year of water), where there is little or no melting or wind scouring, seasonal variations in a number of parameters, including the ratio of the abundance of the stable oxygen isotopes, microparticle concentrations, radioactivity, sodium, sulfate and nitrate ions, have been used to obtain an absolute dating that is accurate to a few years per thousand. Most deep ice cores have produced records that extend back into the last major glaciation, a maximum of 100,000 years before present. An ice core that was recently drilled by the Soviets at Vostok Station, in the central part of the East Antarctic ice sheet, has provided a record that extends back some 150,000 years into the ice age preceding the last interglacial epoch. If it proves possible to drill to the base of the ice sheet at Vostok, the record could be extended perhaps to 500,000 years or more. Although the length of the record that can be obtained from a polar ice core is much shorter than that from deep sea cores (hundreds of thousands of years versus millions of years), the higher accumulation rates in polar ice cores (2 to 50 grams per square centimeter per year versus 1 to 2 grams per square centimeter per thousand years) provide much greater temporal resolution than can be derived from deep sea cores. The determination of the depth-age scale of an ice core is important in properly interpreting other data obtained in the core. As a first approximation, a theoretical depth-age relationship can be calculated if the velocity and annual accumulation rate distributions at the surface of the ice sheet are known (Figure 3, page 59). This theoretical time scale can then be checked with some of the other dating methods, including radiometric dating (gross beta activity, lead-210, tritium, carbon-14, beryllium-10, chlorine-36, silicon- 32), time-stratigraphic horizons (comparison with dated climatic or volcanic events or times of radioactive fallout from bomb tests), and seasonal fluctuations in the concentrations of isotopes (oxygen- 18/oxygen- 16, deuterium/hydrogen) and other impurities (sodium, hydrogen ion, sulfate, and nitrate) in the ice. Temperature and Precipitation. Measurements of the stable isotopes of oxygen and hydrogen (oxygen- 18/oxygen- 16, deuterium/hydrogen) can provide information on paleoclimate and the precipitation history of an area, provided that something is known about the spatial and temporal 56 39 35 31 -13.5 •15 ° -20 c -30 3 0 -40 BYRD CORE -60 -80 LlOO < kilometers Figure 1 . Antarctica, siiowing the location of sites where ice cores have been drilled. Examples of the oxygen-isotope records from the three deep cores — the Vostok core, the Dome C core and the Byrd core — are shown for comparison. Depth and estimated age are shown on the left and right hand side, respectively, of each profile. variations of these parameters at the core site. When local signals are separated out, a record of global climatic variations usually can be obtained. Comparisons with the stable oxygen isotope records from forams in deep sea cores and lake cores that are dated by other techniques also can aid in the temporal correlation of climatic events on a global basis. Studies of Greenland and Antarctic ice cores have shown that climatic conditions, including temperature and precipitation rate, were much different during the last glacial maximum than those at present. Results from the Dome C and Vostok cores, drilled in the central part of East Antarctica, indicate surface temperatures some 8 to 10 degrees Celsius colder during the last glacial maximum and accumulation rates (at least for Dome C) on the order of 25 percent lower than at present. Although the stable-isotope ratios measured in ice cores are usually interpreted in terms of the atmospheric condensation temperature, there are other processes that can affect this record in an ice core, these include changes in the ratio of summer to winter precipitation, and changes in the moisture source of the precipitation as the result of variations in the extent of sea ice or changes in atmospheric circulation. Recent studies by a French group, headed by Jean Jouzel of the Centre d'Etudes Nucleaires at Saclay, France, have led to the definition of a new parameter, the deuterium excess (d), which is calculated by subtracting the oxygen isotope ratio from the hydrogen isotope ratio. The deuterium excess is interpreted in terms of the relative humidity over the oceans. Low values of the deuterium excess observed at the time of the last glacial maximum (about 18,000 years before present) are interpreted to indicate a higher relative humidity over the oceans at that time. Atmospheric Cases. In the processes by which snow transforms into glacier ice, air is trapped in bubbles in the ice. The air that is trapped in these bubbles preserves a record of the composition of the atmosphere at the time of their final isolation. Measurements of the composition and the concentration of these gases provide paleoclimatologists and atmospheric chemists with a window on the history and development of our atmosphere over the last 150,000 years. Of particular interest have been the measurements of carbon dioxide and methane, which have shown significant variations linked with the transition from the last ice age to the Holocene period (about 10,000 years before present). Studies of recent ice from both Antarctica and Greenland have shown that pre-industrial (prior to 1850 A.D.) levels of carbon dioxide were, on 57 CAMP DYE 3 CENTURY kilometers 6 07c 40 35 30 CAMP CENTURY 6 0 7oo Figure 2. Greenland, showing the location o( sites where ice cores have been drilled. Examples of the oxygen-isotope records from two deep cores — Camp Century and Dye 3 — are shown for comparison. Depth and estimated age are shown on the left and right hand side, respectively, of each profile. average, between 260 to 280 parts per million by volume (compared with a present-day concentration of 340 parts per million by volume) — lower than estimates based on subtracting out the recent burning of fossil fuels as a source of carbon dioxide. This lower than expected level suggests the presence of an additional source of carbon, and may be related to the agricultural "revolution," at a time when the slash-and-burn method of subsistence was more common. Furthermore, some variations observed in the period from 1500 to 1900 A.D. suggest that the "pre-industrial" value was itself not constant, and that the natural background preceding the anthropogenic perturbation may have been dependent on climatic fluctuations. Measurements of carbon dioxide also have been made on ice core samples from Greenland and Antarctica that were formed during the last glaciation. These measurements suggest that the levels of carbon dioxide in the atrnosphere during the last Glacial Maximum (the Late Wisconsin advance — about 18,000 years before present) may have been significantly lower than the value estimated for the pre-industrial time (200 versus 260 to 280 parts per million by volume). Rapid increases in carbon dioxide during the transition from the Late Wisconsin Maximum into the Holocene period suggest a link with climate, as measured by the oxygen isotope record of the ice cores. More studies from other deep ice cores are needed, however, in order to confirm this pattern and to establish a precise chronology for this period of time. The question of whether the increased carbon dioxide was a "cause" or a "result" of climatic warming, however, still remains a difficult one. Current research on carbon dioxide in ice cores may soon shed new light on the subject. This puzzle will be an important one to resolve, especially with regard to questions currently being raised over the possible instability of the marine-based West Antarctic ice sheet. In addition to studies of the composition and concentration of gases trapped in the ice, the total gas content of polar ice has been used as an indicator of the surface elevation of the ice sheet at the time that the ice was formed. This information, in conjunction with stable isotope profiles, allows one to correct the temperature record for the effect of increased or decreased elevation. A new approach to the study of ice cores has recently been undertaken by a joint team of French and American researchers, headed by Michael Bender at the Graduate School of Oceanography at the University of Rhode Island. By analyzing the isotopic composition of the oxygen trapped in air bubbles in the ice, changes in the isotopic composition of the past atmosphere have been measured. Bender and his colleagues found that the oxygen isotopic composition of the atmosphere during the last glaciation varied in sympathy with the isotopic composition of seawater (that is, a 1 .3 part per thousand enrichment of the oxygen isotope ratio in both seawater and atmospheric oxygen was measured). They related this to changes in the global rate of primary productivity in the oceans at that time. Atmospheric Aerosols. The history of the atmospheric aerosol* also is preserved in polar ice * A suspension of fine solid or liquid particles in a gas. 58 cores in the form of soluble and insoluble impurities in the ice. The soluble and insoluble impurities preserved in the polar ice sheets originate from a variety of sources, including continental dust, sea salts, biogenic emanations, anthropogenic emissions, extraterrestrial material, and products of atmospheric reactions and volcanic eruptions (both ash and gases). Analyses of the impurities in the ice have been made on a number of cores from both Greenland and Antarctica. These studies have shown that the concentrations of marine-derived aerosols (represented by sodium and chloride concentrations) were about 5 times greater, and that the continentally-derived dust content (represented by aluminum concentrations and silicate particulates) was about 20 times greater, during the last Glacial Maximum than during the Holocene period. This information has led glaciologists and paleoclimatologists to conclude that the climate during the last Glacial Maximum was characterized by enhanced aridity, more vigorous atmospheric circulation, and greater aerosol transport toward the polar regions. Studies of the chemical composition of recent snow and ice in Antarctica and Greenland are currently of interest to researchers who are investigating problems of global pollution and the possibility of climate modification resulting from such pollution. A comparison of the chemical composition of snow layers deposited prior to the industrial revolution with those deposited more recently has allowed an assessment of both the natural background levels of certain chemical species as well as the extent to which the remote regions of the globe have been affected by such pollution. Studies to date have concluded that the concentrations of sulfate and nitrate (derived from anthropogenic sulfuric and nitric acid) have increased threefold and twofold, respectively, in Greenland snow over the last century. Similar increases have not been detected in recent Antarctic snow. Significant increases in the concentration of industrially produced heavy metals (especially lead) have been measured in recent Greenland snow (a 100-fold increase since 800 B.C.). Evidence for an anthropogenic increase of lead in Antarctica is still debated and could range anywhere from no increase to a 40-fold increase (however, smaller values are favored). While results of this kind are not unexpected (at least 90 percent of anthropogenic emissions occur in the Northern Hemisphere), a study completed in 1985 on the record of global pollution in polar ice cores, by Eric Wolff and David Peel of the British Antarctic Survey, suggested that "there are no reliable data for heavy metals in ancient ice in either hemisphere, so anthropogenic trends cannot yet be estimated." A new study, by Claude Boutron of the Laboratoire de Glaciologie in Grenoble, France, and Clair Patterson of the California Institute of Technology, has shown that the concentration of lead in ancient (27,000 to 4,000 years old) Antarctic ice fluctuated in response to changing levels of continental dust (a 30-fold increase in the aluminum DISTANCE FROM ICE DIVIDE /kilometers Figure 3. Ice flow pattern predicted by a theoretical flow model. Thin curved lines show particle paths (trajectories that a particle would take). Dotted lines are isochrones (equal time lines), with ages in thousands of years marked on the right hand side. (After Reeh, and others, 1985). concentration from continental dust produced a 30- fold increase in the lead concentration). Volcanoes and sea salt were not found to be significant sources of lead to the atmosphere during the last glaciation. However, when continental dust levels were lower, during the Holocene, the natural flux from volcanoes made a more significant contribution to the atmosphere (50 percent of the excess lead in the Holocene ice was from volcanoes). Boutron and Patterson's results suggest that more than 99 percent of the lead now in the troposphere* comes from human sources and that, at least for lead, there are few, if any, natural sources currently contributing significantly to the atmosphere Dynamics of the Ice Sheet. Finally, ice cores provide the necessary samples required for studies of the physical properties of glacier ice, including both its mechanical and electrical behavior. When combined with in-situ measurements made on the ice sheets of the densification and recrystallization of the ice, its velocity (basal sliding versus rate of internal, vertical, and horizontal deformation), and temperature variations with depth, these data provide the information necessary for modelling the flow characteristics (or rheology) of the ice. Such studies are necessary in order to determine the present and past flow patterns of the ice sheets, including changes in thickness. This information is extremely useful for interpreting the ice core records and is necessary to establish a theoretical depth-age relationship for an ice core. Models of glacier flow that incorporate information on the physical properties of the ice are also important for climatologists wishing to predict the likely response of glaciers and ice sheets to climatological perturbations. Changes in temperature, precipitation, and atmospheric circulation patterns can all have an affect on the behavior of large ice masses. Complicated feedback mechanisms must also be accounted for when trying to predict whether a glacier will advance, recede, surge (for example the Hubbard Glacier in Alaska) or remain the same, as a result of these changes. * The portion of the atmosphere next to the Earth's surface out to a distance of 7 to 10 miles. 59 Antarctic and Greenland Ice Cores The development of the modern methods and techniques for the retrieval and study of ice cores from polar glaciers was initiated and promoted during the International Geophysical Year (ICY) (1957-1958), and provided the foundation for the science of polar glaciology as we know it today. The study of ice cores from the two great ice sheets of Antarctica and Greenland has proven to be both interdisciplinary and international in nature. Efforts to retrieve and study polar ice cores over the last 30 years have combined the logistical support, technical skills, and scientific expertise of specialists in a number of countries. Figure 1 , page 57, is a map of Antarctica showing the sites where shallow (less than 100 meters), intermediate (more than 100 meters) and deep (more than 500 meters) ice cores have been drilled since the IGY. Examples of the oxygen- isotope records from the three deep cores — the Vostok core, the Dome C core, and the Byrd core — are shown for comparison. The 2,164-meter-long ice core drilled at Byrd Station in 1968 is the only deep drilling that has ever fully penetrated the Antarctic ice sheet. The Byrd core provided the first significant record of climate and atmospheric composition (including insoluble particulates, aerosols, and gases) for the southern hemisphere, as well as an important record of volcanic activity from local volcanoes in West Antarctica. The 2,083-meter core, drilled by the Soviets in 1984 at Vostok Station, where the ice is more than 3,500-meters thick, contains the oldest dated record obtained on any ice core. It is estimated to cover the last 150,000 years, into the ice age preceding the last interglacial epoch. Studies on the Vostok core have already provided an important time series, which is proving useful both for comparison with climatic records obtained from deep sea cores and for testing models which favor the role of orbital forcing to explain climate changes. Figure 2, page 58, shows a similar map of the Greenland ice sheet with the location of sites where ice cores have been drilled since the IGY. A figure showing the oxygen-isotope profiles produced from the analysis of the deep ice cores from Camp Century and Dye 3 is also shown, for comparison with the Antarctic oxygen-isotope records. The work in Greenland has been carried out mainly under the auspices of the Greenland Ice Sheet Program (GISP), which was a collaborative effort begun in the early 1970s between the United States, Denmark, and Switzerland. Prior to GISP several ice cores had been drilled in Greenland by the United States, including the first ever core to bedrock on a polar ice sheet, drilled at Camp Century in 1966, to a depth of 1,385 meters. In 1981, the second deep ice core to bedrock in Greenland (2,037 meters long) was accomplished at Dye 3 station in south Greenland. An excellent quality core was recovered in this drilling effort and a complete climatic record back some 70,000 years or more was produced from studies of the core. Comparison of the oxygen-isotope record obtained from the Dye-3 core with that of the Camp Century core discloses important similarities, confirming the climatic significance and regional importance of the data. Interhemispheric comparisons using ice core records from Antarctica, as well as climatic records from deep sea cores, have provided cross-checks on the dating accuracy of the cores, allowing global climatic oscillations to be described and speculations as to their origins discussed. Finally, complimentary studies of a number of parameters, including atmospheric gases, aerosols, dust content, and radioisotopes, to name a few, have been accomplished, revealing broadly similar trends to those observed in the Antarctic cores. Future studies on these and other new cores from Greenland should provide the information necessary to understand the interhemispheric relationships and the climatic significance of the data obtained on these cores. Integration of Data Needed Polar ice sheets are excellent storehouses of information for deciphering the history of our global atmosphere and paleoclimate. Ice cores provide information on past temperature, precipitation, atmospheric composition, and the physical state of the ice sheet. Long records extending back several hundreds of thousands of years may eventually be recovered from the polar ice sheets; at present the oldest record, from Vostok Station in central East Antarctica, extends back 150,000 years into ice formed during the ice age preceding the last interglacial epoch. This ice core, along with others from Antarctica and Greenland, is providing paleoclimatologists and atmospheric modellers with the information necessary to understand the underlying causes of climatic changes. In the future, it will be important to integrate ice core studies with other information from deep sea drilling cores and more sophisticated atmospheric-climatic modeling. To obtain a true "global" perspective, interhemispheric comparisons of the ice core records from Antarctica and Greenland should be made and integrated with the high-resolution proxy records currently being obtained from tropical and temperate latitude ice masses. Comprehensive studies of this kind should help in the effort to resolve questions currently being posed on problems as diverse as El Niino, global pollution, atmospheric carbon dioxide, and the fate of the West Antarctic ice sheet. lulie M. Palais has participated in three field expeditions to Antarctica. She is a Marine Research Associate at the Graduate School of Oceanography, University of Rhode Island. Suggested Readings Bender, M., L. D. Labeyrie, D. Raynaud, and C. Lorius. 1985. Isotopic composition of atmospheric O2 in ice linked with deglaciation and global primary productivity. Nature 318: 349- 352. Dansgaard, W. 1981. Ice core studies: dating the past to find the future. Nature 290: 360-361. Wolff, E. W., and D. A. Peel. 1985. The record of global pollution in polar snow and ice. Nature 313: 535-540, 60 POLLUJTn©! BULLETHKf Marine Pollution Bulletin is concerned with the rational use of maritime and marine resources in estuaries, the seas and oceans. A wide range of topics are discussed, as news, comment, reviews and research reports, not only on effluent disposal and pollution control but also on the management and productivity of the marine environment in general. The Bulletin also provides information and comment on events with implications for the human use and enjoyment of the seas and coastal environment. First published in 1970, it has proved to be an important and influential journal. New Patents Section — The journal now contains abstracts and illustrations of recently issued United States Patents and published patent applications filed from over 30 countries under the Patent Co- operation Treaty. Software Survey Section — This new section reports developments in appropriate specialist software. Subscription Information Published monthly (Volume 17) Annual subscription (1986) Two-year rate (1986 87) US$85 00 US$161 50 MARINE POLLUTION BULLETIN The International Journal for Marine Environmentalists, Scientists, Engineers, Administrators, Politicians and Lawyers Editor: R B CLARK, Department of Zoology, The University, Newcastle-upon- Tyne NE1 7RU, UK A selection of papers Roles of the oceans in the CO2 question, A J CRANE&PS LISS. Oiled Magellanic penguins in Gulfo San Jose, Argentina, J PERKINS. Shell thickening in Crassostrea gigas: organotin antifouling or sediment induced? M J WALDOCK & J E THAIN. Aerial flux of particulate hydrocarbons to the Chesapeake Bay estuary, D B WEBBER. A history of metal pollution in the Upper Gulf of Thailand, M HUNGSPREUGS& C YUANGTHONG. Effects of metal on sea urchins development — a rapid bioassay, H H LEE &C H XU. Comparative environmental chemistries of metals and metalloids (viewpoint), E D GOLDBERG. Marine pollution research facilities in the People's Republic of China (viewpoint), DA WOLFE ef a/. Estimates of oil concentrations in Aegean waters (baseline), G P GABRIELIDES etal. The influence of experimental sewage pollution on the lagoon phytoplankton, N FANUKO. Reef-building coral skeletons as chemical pollution (phosphorus) indicators, R E DODGE era/. FREE SAMPLE COPIES AVAILABLE ON REQUEST Advertising rate card available on request Back issues and current subscriptions are also available in microform Prices are subject to change without notice. Journal prices include postage and insurance Sterling prices are available to UK and Eire customers on request Pergamon Press Headington Hill Hall, Oxford 0X3 OBW, UK Fairvlew Park, Elmsford, New York 10523, USA 61 Humans in Ice Age Europe- Bison licking its flank, sculpted from reindeer antler (part of spear). From La Madeleine, France; Middle Magdalenian (17,000 to 14,000 B.C.). Approximately 20,000 years ago. Upper Paleolithic peo- ple invented the spear thrower to extend hunting capacity. This, and the other ar- tifacts on these pages are part of a dis- play at the American Museum of Natural tiistory in New York City until January 18, 1987, entitled "Dark Caves, Bright Visions: Life in Ice Age Europe. " (Photo by Musee des Antiquites Nationales.) Lamp of red sandstone. A long incision and several shorter strokes decorate the top of the long handle. From Lascaux Cave, France; Early Magdalenian (20,000 to 17,000 B.C.). The lamps exhibited re- flect how humans controlled heat and light during the late Ice Age. Several lamps made of limestone were hollowed in the center to hold fuel. (Photo by Mu- see des Antiquites Nationales.) Fiorse sculpted in the round in ivory. The neck is horizontal, and the legs, with lower parts broken, are separated from each other. Details of the coat are indicated only on one side. From Lourdes, France; Middle Magdalenian. Although the culture of the late Ice Age is usually symbolized by famous cave paintings, portable art objects, such as this horse. Illustrate the burst of culture and art that occurred among Upper Pa- leolithic Europeans between 35,000 and 10,000 years ago. (Photo by Musee des Antiquites Nationales.) 0,000 to 35,000 Years Ago Necklace composed of pierced teeth 3nd shells. The (our large teeth are of bear and lion. The lion tooth has been incised on all sides. From Rocher de la Peine, France: Late Magdalenian (14,000 to 1 1,000 B.C.). Anthropologists assume that such bodily adornment communi- lated social position. (Photo by R. White.) The mind is the form of forms, and the hand is the tool of tools. When the hand and mind un- ite to forge objects of beauty and or- der, these may be works of science or of art — Aristotle. Two headless ibexes (mountain goats) embracing, playing, or fighting. Carved Irom reindeer antler for spear. The hook, bottom left, was inserted into the end of the spear shaft. From Fnlene Cave, hance; Middle Magdalenian. More so- phisticated technology, such as the use of chisels, was used to produce art dur- ing this period than by the Neander- thals, who were the predecessors of Ice Age humans. Many people think of the last major Ice Age as having occurred (luring a period of interest only to pa- kontologists or geologists. But, as we see here, very sophisticated humans were already abroad in the world. Fngraving of a horse's head on a flat piece of bone. Narrow incised lines dec- orated with oblique incisions and run- ning from muzzle to ear are considered by some to be part of the halter. From St. Michel-d'Arudy, France; Middle Mag- dalenian. The functions of many of the objects remain unknown. (Photo by Mu- see des Antiquites Nationales.) Pollen In Marine Cores: by Linda E. Heusser I o look in the oceans for direct evidence of past continental climates seems paradoxical. However, marine sediments contain far better terrestrial paleoclimate data than most continental deposits. On land, records of past climates found in glacial deposits, tree rings, or lakes are limited spatially and temporally, and prior to about 50,000 years ago accurate dating information is lacking. Although we are increasingly concerned with the relationship between the terrestrial biosphere and future global climate changes — the effect of increased carbon dioxide or a "nuclear winter" — our limited knowledge of past continental environments constrains predictions of future climates and environments. To predict the future, we need to understand the past. The major source of global climate information lies in marine sediments. Cores from the world's oceans provide continuous, lengthy, chronologically-controlled climate-related signals, such as global ice volume and sea-surface temperatures. Deep sea cores from continental margins — the transition zone between oceanic and continental realms — also contain terrestrial climate signals: pollen derived from vegetation growing onshore. Vegetation and Climate The close relationship between the distribution of vegetation and climatic variables, observed by early 64 Evidence Of Past Climates naturalists such as Charles Darwin and Alexander von Humboldt, forms the basis of using changes in plant abundance and distribution as a source of terrestrial paleoclimatic information, particularly during the last million years when evolution and extinction of plant species are minimal. Historically- documented changes in plant distribution, such as the disappearance of beech and chestnut trees from Rome in the first century, and variations in tree growth during the last thousand years, are closely related to temperature fluctuations. For longer vegetation records, the best source is pollen. Pollen As a Vegetation and Climate Signal As hay-fever sufferers know, pollen is ubiquitous and abundant. Deposited and preserved in lakes and bogs, these microscopic grains (Figure 1) document the nature of the surrounding vegetation. Fossil pollen preserved in lake deposits provide an estimate of environmental changes reflected by changes in plant communities around the lake. Quantitative climatic reconstructions are derived by calibrating spatial correlations between modern climate variables, such as temperature or precipitation, and pollen data sets, using the same Above, Figure 1 . Examples of terrestrial pollen typically incorporated into marine sediments. The diameter of these pollen grains is on the order of 20 to 1 20 microns. 65 Figure 2. Most of the millions of pollen grains dispersed into the atmosphere are deposited within a short distance (less than a thousand meters) from their source, as symbolized by the left- hand arrow showing pollen from a lone pine tree dropping into a nearby lake. Pollen from vegetation growing immediately along the seacoast is carried by wind to the ocean, as shown by the right-hand arrow. Empirical studies show that fluvial transport (via streams and rivers), the middle arrow, carries most pollen to the ocean, where it is deposited along with the carbonate skeletons of marine microorganisms. Deep sea cores from the continental margins subsequently retrieve sediment containing both marine microorganisms and pollen — an unequaled source of terrestrial and marine paleoclimatic data. multivariate statistical techniques pioneered by John Imbrie and N. G. Kipp to transform marine plankton data into quantitative estimates of sea- surface characteristics. Most Quaternary* pollen records are less than 25,000 years old. Although these records provide detailed information about high-frequency climatic events since the end of the last major glaciation, long climatic signals, on the order of hundreds of thousands of years, are needed to understand the role of the terrestrial biosphere in the climate system. These signals are contained in ocean sediments. Pollen in Marine Sediments Pollen is abundant in marine sediments deposited on continental margins (the transition between the continents and the deep ocean basins). As on land, in ocean sediments pollen reflects the environmental parameters of the vegetation from which it is derived. Carried by wind and rivers to the ocean (Figure 2), pollen patterns in marine sediments generally correspond with regional vegetation patterns on land. In the northeast Pacific Ocean, distribution of coastal redwood pollen (Sequoia sempervirens) is basically restricted to sediments deposited off the redwood groves of northern California, and hemlock (Tsuga) pollen characterizes the sediments adjoining the magnificent hemlock- dominated conifer forests of coastal Washington and Oregon. In Japan, spruce (Picea) pollen dominates marine sediments surrounding the boreal forests of * The Geological Period extending from 1.6 million years before present to the present. Hokkaido in the north. Pollen from the warm temperate forests of southern Japan, such as Japanese cedar (Cryptomeria) is prominent in sediments close offshore (Figure 3). This systematic relationship between pollen distribution in marine sediment and vegetation onshore is the basis for the calibration between marine pollen and continental environmental parameters. Deep Sea Pollen Records Deep sea pollen records provide a unique means of directly relating regional continental climate records with regional and global marine records because deposition of pollen grains in ocean muds corresponds with deposition of other components of the same sediment sample. Therefore, marine pollen records are correlated directly with the marine microfossil records from the same core. Thus, the pollen records are precisely related to global timescales developed from these marine microfossils. The data from marine cores in different ocean basins differ both in length and in the climatic signal provided. Temperature and precipitation changes in northwestern Europe are correlated with global ice volume changes over a 50,000 year time span. Marine pollen in cores from the northeastern Pacific link the climatic history of northwestern North America to regional and global changes in the oceans and the ice sheets over the last 1 30,000 years. Paleoclimatic records of similar length from the Arabian Sea and the northwest Pacific document correlative continental and marine climatic fluctuations since the end of the previous ice age (140,000 years before present). These include changes in monsoonal wind intensity, precipitation, and temperature. Several examples follow. 66 Northeast Atlantic Ocean. Pollen records from a core taken in the northeast Atlantic Ocean about 100 kilometers off the Iberian Coast, and a core from northeastern France link western European climatic sequences from 125,000 to 75,000 years before present with global climate changes. French scientists conclude that during this time, the sequence of northwest European climatic events inferred from the pollen data generally agrees with changes in ice volume suggested by the oxygen isotope* curve. During the last interglacial, ice volume apparently increased before climatic deterioration occurred in southwestern Europe. Full glacial conditions on land began after major ice-sheet accumulation. European climatic changes inferred from pollen signals agree with paleotemperature trends in the northeastern Atlantic Ocean. Both the maritime climates of western Europe and waters offshore remained warm during the initial phases of high-latitude ice-sheet accumulation at the beginning of the last glacial cycle. Northeast Pacific Ocean. The first continuously-dated history of terrestrial climatic change in northwestern North America over the last 150,000 years is reconstructed from pollen and oxygen isostope analyses of cores taken in the northeast Pacific Ocean (Figure 4, page 68). Pollen profiles show nonglacial intervals of coastal lowland forest (typified by western hemlock) alternating with glacial intervals in which spruce and herbs are relatively more important. Temperatures, inferred from the ratio of western hemlock to spruce pollen, are highest during interglacials (isotope stage 1 and substage 5e) and lower during glacial events (isotope stages 2, 4, and 6). Floral and faunal data from this deep sea core show similar terrestrial and marine climatic trends. Temperate forest pollen is highly correlated with the presence of subtropical and transition zone radiolarian faunas and with the absence of subpolar fauna. The herb-dominated pollen assemblage, prominent during glacial intervals, is correlated with the subpolar radiolarian assemblage found during intervals of increased dominance of marine subpolar conditions in the northeast Pacific Ocean. Mathematical analysis of oxygen isotope, radiolarian (siliceous marine microfossils), and pollen data by Nicklas Pisias of Oregon State University indicates that most variations in these marine and land signals are found at periods around 41,000 years, the period of the orbital tilt cycle (see page 43). The occurrence of major changes in the vegetation of northwest North America at periods characteristic of orbitally-controlled insolation changes provides empirical evidence relating terrestrial climatic changes to orbital forcing. iPANESE CEDAR (Cry ptomeria ) 140 160 Figure 3. The geographic distribution of two diagnostic pollen genera in marine sediments of the northwest Pacific Ocean. Areas with significant amounts of pollen are shaded, with darker tones indicating the highest percentages of pollen. Spruce pollen (a) is most abundant near the spruce- dominated boreal forests of eastern Siberia, southern Sakhalin, and Hokkaido, the northernmost island of the Japanese archipelago. The highest quantities of Japanese cedar pollen (b), an endemic component of the warm temperature forests of Japan, are found in the marine sediments adjoining these forests in southern central japan. * The ratio of oxygen- 18 to oxygen- 16 is commonly used in strafigraphic analysis. Oxygen consists mainly of oxygen-16 atoms, mixed with a relatively few oxygen-18 atoms. The proportion ot the heavier oxygen-18 in the molecules of ocean water changes when the climate changes, since molecules containing the lighter atoms evaporate more readily. When they fall as snow, and become locked up in ice sheets, they leave the oceans relatively enriched in oxygen-18 — a sign of glacial conditions. These isotopes are recorded from the shell fossils of small marine animals present at the time. When extracted from a core sample of seabed sediment, a handful of such fossils is sufficient to determine the volume of the world's ice sheets at the time when they lived. 67 Temperature Cool ► Warm Ice Sheet Volume Low ► High Last Glacial Maximum Last Interglacial 4 5a 5c 5e Western Hemlock / Spruce Ratio 6''o Figure 4. Climate signals from the last 150,000 years from northwest North America and the northeast Pacific Ocean, as obtained from a 16-meter deep sea core. Paleotemperature trends for coastal Washington and Oregon (left) are inferred from the ratio of western hemlock to spruce pollen (temperatures in western hemlock-dominated coastal forests are I to 2 degrees higher than in spruce- dominated forests). Global ice- sheet volume (right) is derived from changes in the oxygen isotope composition of the calcareous tests of foraminifera contained in the same sediment sample as the pollen. Cross-spectral analysis (mathematical comparison of signal frequency) shows that environmental changes on the Pacific coast of North America are not precisely synchronous with temperature changes in the North Pacific Ocean, or in global ice volume. Temperate conditions onshore are established after temperate conditions offshore, and changes in ice volume in the subarctic Pacific slightly precede sea-surface temperature changes. Arabian Sea. American and French scientists (Warren Prell, Brown University, and Elise van Campo, University of Languedoc) reconstructed monsoonal wind intensity over the last 130,000 years from pollen and foraminifera assemblages in a core from the western Arabian Sea. Seasonal differences in the composition of aeolian (wind-borne) pollen in the Arabian Sea are related to the seasonal reversal of winds. Increased amounts of diagnostic pollen types from humid or montane areas of northeast Africa and southwest Arabia are associated with increased summer monsoonal winds and increased precipitation. Increases in the planktonic foraminifera (calcareous marine microfossils) signal, which is presently associated with temperature and nutrient evidence of summer coastal upwelling, are interpreted as evidence of stronger winds during the summer southwesterly monsoons. Intensified summer southwest monsoons during interglacials are inferred from pollen and foraminifera. Both proxy climate indicators are coherent and in phase at 23,000 years, a period associated with the precessional cycle of the Earth's orbit (see page 43). Empirical evidence from Africa and the Arabian Sea show changes in the Indian summer monsoon associated with changes in seasonal solar radiation. Northwest Pacific Ocean. The complexity of climatic change during the last 150,000 years is illustrated by terrestrial and marine signals from deep sea cores taken along a south-north transect in the northwest Pacific Ocean off the coast of Japan, in recent work done by J. Morley and myself at the Lamont-Doherty Geological Observatory of Columbia University. The first environmental records from the Pacific Coast of Japan covering the last interglacial-glacial cycle are represented by pollen assemblages (Cryptomeria, Quercus, Pinus, and Picea). Sediments in core RC14-99 record major changes between warm and cool temperate vegetation, 70,000 to 128,000 years ago. Warm temperate vegetation, characterized by the Japanese cedar (Cryptomeria) time series, expands during nonglacial intervals. Glacials, on the other hand, are characterized by expanded cool temperate oak (Quercus) forests, spruce (Picea) forests, and by the presence of an arctic/alpine indicator species (Figure 5, page 69). The last interglacial, clearly identified by the expansion of Japanese cedar between approximately 125,000 and 1 16,000 years before present is succeeded by two subsequent episodes of warm temperate forest expansion prior to the onset of full glacial conditions about 70,000 years ago. Pollen profiles from a core, located near the warm-temperate vegetation of southernmost Japan, also document expanded Japanese cedar-dominated forests prior to 70,000 years ago. However, regional environments differed substantially during the last interglacial (oxygen isotope substage 5e). On the Pacific coast, Japanese cedar was strikingly unimportant in forests of southern Japan, while dominating forest communities in central Japan. These climatic trends are summarized in Figure 6, page 70. The precipitation and temperature indices from the Pacific coast of Japan essentially reflect the importance of the standard Japanese paleoclimatic indicators of Japanese cedar, hemlock 68 (Tsuga), and spruce. Summer temperature fluctuations interred from pollen data in cores from 36 and 40 degrees North appear to correspond with regional Japanese paleoclimatic reconstructions, and with global ice volume changes and global estimates of sea-surface temperatures. Temperature estimates from central Japan differ markedly from summer sea-surface temperatures offshore (at the core site) during the last interglacial, as do temperature estimates for southern Japan represented in the core from 28 degrees north. Like temperature, precipitation indicators from southern and central Japan show different trends in the interval between 140,000 and 120,000 years ago, suggesting drier, cooler conditions in the southeast as compared with the central coast. Climatic signals from the vegetation of the Pacific coast of the Japanese archipelago over the last 150,000 years reflect global, regional, and local climatic variations with differing sensitivity. The transition or tension zone vegetation of central Japan appears closely tuned to global variations related to orbital forcing, as expressed in classic ice volume curves. Large-scale circulation changes associated with the Asian monsoon are suggested as important factors in climatic variations in southernmost Japan. In the last interglacial, for example, a stronger summer southeastern monsoon associated with northward movement of the atmospheric front would be consistent with our terrestrial temperature and precipitation reconstructions. Low glacial temperatures inferred from floral and faunal assemblages probably reflect intensified winter Figure 5. A scanning electron micropiiotograph (SEM) of a Japanese arctic/alpine indicator species, the spiked moss, Selaginella selaginoides. The diameter of this pollen grain is about 50 microns. monsoons, as well as effects of Northern Hemisphere glaciation on regional atmospheric circulation. J^.t~ A woodblock landscape from a series of 53 stages on the royal road from Edo (Tokyo) to Kyoto (the former capital), by Ando hiiroshige (1797-1858). Pollen from Japanese trees, a record of climate, is archived in coastal marine sediments. 69 Ice Sheet Volume (from oxygen isotopes ) Summer Sea- Surface Temperature N.W. Pacific Precipitation Japan Temperature Japan Last Glacial Maximum Last Interglaciat Low — ^ High — ► Warm — Figure 6. Northeast Asian paleoclimatic signals from the last 140,000 years. These climate records are derived from climatically- sensitive marine microfossils (selected foraminifera, radiolaria, and pollen) in cores taken east of lapan. Precipitation trends reflect the importance of Japanese cedar, and temperature trends reflect the prominence of spruce-dominated boreal forests in japan. Except for the last interglacial v^hen sea-surface temperatures were low, marine temperature fluctuations mirror changes in the vegetation and climate of north-central japan. Pollen Analysis Vital Over the last 150,000 years, changes in broad-scale vegetation patterns, as documented by marine pollen, are correlated with changes in global temperature and precipitation. The frequency of these past fluctuations is associated with periodic changes in solar insolation. Future changes in components of the climate systems — variations in solar insolation or increased atmospheric carbon dioxide — will dramatically change the environment of the land we live on. How it will change is not precisely known. However, extending terrestrial climatic records in time and space is fundamental to understanding both the past and the future of the climate system. Cores, particularly those on continental margins taken by the Ocean Drilling Program and by oceanographic institutions, such as the Lamont- Doherty Geological Observatory and the Woods Hole Oceanographic Institution, will extend records backward through time, and toward worldwide coverage. Pollen analysis from these cores will provide a worldwide terrestrial climate data bank. From the analyses of these data we may be able to better predict changes in both species composition and range of the vitally-important terrestrial flora as climate conditions on our planet continue to change. Linda E. Heusser is a Research Scientist at the Lamont- Doherty Geological Observatory of Columbia University, Palisades, New York. Selected References Heusser, L. E., and N. ). Shackleton. 1979. Direct marine-continental correlation: 1 50,000-year oxygen isotope-pollen record from the North Pacific. Science 204: 837-839. Heusser, L. E., and ). |. Morley. 1985. Pollen and radiolarian record from deep-sea core RC14-103: climatic reconstructions of northeast japan and northwest Pacific for the last 90,000 years. Quaternary Research 24: 60-72. Imbrie, )., and N. G. Kipp. 1971. A new micropaleontological method for quantitative paleoclimatology: application to a Late Pleistocene Caribbean core. In The Late Cenozoic Glacial Ages, ed. K. T. Turekian, pp. 71-182. New Haven, Connecticut: Yale University Press. Molfino, B., L. E. Heusser, and G. M. Woillard. 1984. Frequency components of a Grande Pile Pollen record: evidence of precessional orbital forcing. In Milankovitch and Climate, Part /, eds. A. L. Berger, ). Imbrie, J. Hays, G. Kukia, and B. Saltzman, pp. 391-404. Boston: D. Reidel Publishing Co. Prell, W. L., and E. Van Campo. 1986. Coherent response of Arabian Sea upwelling and pollen transport to late Quaternary monsoonal winds. Nature 323: 526-528. Turon, |. L. 1984. Direct land/sea correlations in the last interglacial complex. Nature 309: 673-676. Van Campo, E., |. C. Duplessy, and M. Rossignol-Strick. Climatic conditions deduced from a 1 50-kyr oxygen isotope-pollen record from the Arabian Sea. Nature 296: 56-59. 70 t m'' 1» *' fe&K* 'i lal iwrn: :' V .t ■ WW '^'7 ^-- ja>^ 3 ' ^'^''''^W^^H / ^M^ ^" A fc'lk.^'^- .1>J ' ^..- c -.1,1 .:i^f^''- m0<^ ^*^ i*"^^. T-^'^'^ 4. Forests and Climate: Surprises in Store by George M. Woodwell If the climatic changes anticipated from the increase in carbon dioxide and other trace gases in the atmosphere follow predictions made in the fields of meteorology and oceanography, the shocks to contemporary civilization will be substantial. The analyses predict the beginning of an indefinite period of global warming. The warming will be greatest near the poles; there will be little change in the tropics. The analyses have been based on physical and chemical data. However, they have included only superficial consideration of the biotic interactions that affect the atmosphere. The biotic considerations are important because they throw long-standing assumptions into question, and show that the climatic changes are likely to be more serious and threatening than commonly envisioned. Measurements in 1986 by P. D. Jones and others as reported in the British magazine Nature suggest that the warming trend is now under way. The carbon dioxide content of the atmosphere has been rising throughout the last century. The amount in the atmosphere is now about 350 parts per million by volume, 25 to 30 percent above the amount present a century ago. The cause of the increase is generally accepted as the combustion of fossil fuels. However, evidence suggests that the destruction of forests over the last century and a half was the major source of carbon dioxide released into the atmosphere until the middle 1960s. The current release of carbon into the atmosphere from the combustion of fossil fuels is about 5 X 10'^ grams annually. There is further release from deforestation that is difficult to measure. Recent estimates suggest that this release is in the range of 0.5-4.7 X 10'^ grams. The total amount of carbon released annually into the atmosphere from these two sources probably lies in the range of 6-9 x 10^^ grams, (6 to 9 billion metric tons) but the actual amount is not known. The atmospheric concentration is easily observed. It is rising at about 1.5 parts per million, equivalent to 3.0 x 10^^ grams annually. The difference between the total amount emitted into the atmosphere and the amount that accumulates in the atmosphere is absorbed into the oceans or into the terrestrial biota and soils. These views are widely accepted, at least up to the point where there is a significant additional release of carbon from the terrestrial biota into the atmosphere. But studies of deforestation show that this additional increment must be added to the release from fossil fuels to determine the total effect of human activities. Studies of oceanic uptake of carbon dioxide, however, seem to show that the amount absorbed by the oceans is limited to less than the difference between the amount accumulating in the atmosphere and the total released from fossil fuels. That is, some of the CO2 from the burning of fossil fuels cannot be accounted for in our budgets, let alone the excess CO2 expected because of deforestation. If this is correct, the forests are either absorbing carbon from the atmosphere, or, releasing an amount that is equal to a very small fraction of the fossil fuel release. For this reason a mechanism has been sought that would cause the biota to store an increasing amount of carbon as the atmospheric concentration increases. The Role of the Biota The issues become more complicated, confusing, and threatening when the potential influence of the biota is considered more carefully. The amount of carbon held in reservoirs that are directly under biotic control is large by comparison with the amount held in the atmosphere. These large, biotically-controlled reservoirs are on land, although the potential of planktonic populations in the sea for sequestering carbon cannot be overlooked. The largest reservoirs of carbon compounds on land are in the plants and soils of forests. The total amount held in plants globally is in the range of 500 to 700 X 10^^ grams, or more; the amount held in soils is uncertain, but commonly thought to be about 1,500 x 10^^ grams. These two reservoirs contain about 3 times the amount held in the atmosphere. A small change in the metabolic processes that affect the size of the terrestrial pools has the potential for affecting the atmospheric composition appreciably. The importance of these biotic processes is emphasized in the sinusoidal annual oscillation conspicuous in the data from monitoring stations of the middle and high latitudes in the Northern Hemisphere (page 9). The oscillation is now widely accepted as caused by the annual cycle of metabolism of seasonal forests. Their metabolism is dominated in spring and summer by net photosynthesis that results in a net storage of carbon in plants and soils. Consequently, the concentration of CO2 in the atmosphere is reduced. During fall and winter, on the other hand, respiration dominates and there is a net release of stored carbon into the atmosphere. Obviously, any small change in the rate of one of these processes has the potential for affecting the composition of the atmosphere significantly in a few weeks. The question is, what can be anticipated as the composition of the atmosphere changes and the global climate warms? The Mauna Loa Record Recently a consistent year-by-year increase in the amplitude of the oscillation observed at the Mauna Loa station in Hawaii has been measured. The increase was assumed immediately by some to be the result of a stimulation of carbon fixation by the increase in the concentration of carbon dioxide in the atmosphere. They reasoned that because carbon dioxide is required for photosynthesis, an increase in the concentration of CO2 in the atmosphere would increase the rate of carbon fixation and the total amount of carbon removed by the biota seasonally. Such an effect would be consistent with current models based on limited absorption by the oceans. The argument is supported by abundant experimental evidence showing that increased concentrations of carbon dioxide in air increase rates of growth of plants in agriculture. There is, however, no evidence from natural populations that the 25 to 30 percent increase in carbon dioxide in the atmosphere over the last century has 72 increased the general growth of trees or other perennials. Even in agriculture where other factors, such as water and nutrients, can be kept available, the stimulation of growth from a 25 to 30 percent increase in carbon dioxide is small, and probably not more than a few percent under the best circumstances. But quite apart from the naivete inherent in accepting a feebly supported assumption, the amplitude of the oscillation is obviously the sum of photosynthesis and respiration over some segment of the hemisphere. Is it possible that respiration also might be stimulated in some way, generally or seasonally? What are the factors that affect the ratio of gross photosynthesis to total respiration in terrestrial ecosystems? Terrestrial Metabolism It is, of course, difficult to measure the metabolism of terrestrial ecosystems as units and even more difficult to experiment with the factors that may control their metabolism. Measurements have been made, however, under limited sets of circumstances. It is also possible to infer from more narrowly focused physiological experiments how forests might be expected to behave as physical conditions change. In any review of the list of factors that affect photosynthesis and respiration, and might be important in shifting the ratio of these processes, the availability of energy, water, mineral nutrient elements, and changes in temperature all have large effects. The availability of water and nutrients, for example, commonly affects production in agriculture. The concentration of carbon dioxide appears in this list, but far down, and its effect is small by comparison with effects expected from other factors. Temperature, on the other hand, may have profound effects on rates of respiration, but very little direct effect on rates of photosynthesis. A 10- degree Celsius increase in temperature close to the normal temperature for a species commonly induces a change in the rate of respiration of 1.5 times to more than 4 times the previous rate. Similar relationships exist for organic soils in the Arctic and probably for terrestrial ecosystems in general, but because of limited evidence, the question is far from resolved, despite its potential importance in determining climate. The data available from studies of ecosystems are limited because research on ecosystems as units has been limited. In one program that did take place, a series of efforts was made at Brookhaven National Laboratory in the 1960s to examine the metabolism of forests. Various techniques were used, including an array of specially-designed chambers. The underlying principle was to measure continuously the amount of carbon dioxide absorbed or emitted over time as an index of metabolism, net photosynthesis in green tissues during daylight, and respiration. In one study, nocturnal temperature increases resulted in the accumulation of carbon dioxide within a forest over a period of several hours. The rate of increase in the concentration of carbon dioxide was assumed to be correlated with its rate of emission and with the rate of respiration of the ecosystem. Data were taken for more than 40 such inversions over the course of one year. Correlations with temperature were observable with different curves defined for winter and summer. The change in the rate of respiration for a 10-degree Celsius change in temperature is referred to by physiologists as a Q10. A Q10 of 2 means that a 10 degree Celsius increase in temperature produces a doubling of the rate of respiration. In the data from the forest of central Long Island, the QIOs for the separate curves defining respiration for winter and summer were similar, 1.3 to 1.5, although the rates at the same temperature differed appreciably between winter and summer. A Q10 calculated between winter and summer was much higher, about 3.5. The experience is much too limited to be used widely in calculations such as these. It is being supplemented now by studies of the metabolism of arctic tundra, but there are few data that define forest respiration rates as a function of environmental factors. The most reasonable assumption at the moment seems to be that the QIO for forest respiration is of the order of 1.3 to more than 3.0. That means that an increase in temperature of 1 degree Celsius can be expected to increase the rate of forest respiration as a whole by between 3 and 25 percent. A warming in the middle to high latitudes is expected over the next one to three decades of several degrees Celsius. The warming could easily double the rates of respiration in the current biota, including soils. Is there a parallel effect on photosynthesis that can be expected to compensate for such a stimulation of respiration in forests? Climatic Change and Forests Ecologists have little difficulty in proving the migration of forests as climates have changed in the past. The forests of New England have all developed on land churned by glacial activity within the last 15,000 years and there is evidence that the adjustments of natural communities are still under way. Treelines migrate, species migrate, and soils are built and destroyed as physical and chemical circumstances change. But, the response times for forests are commonly measured in decades to centuries, not years to decades. Sudden changes in climate, or in other conditions, bring the devastation of forests that we are seeing now in Europe and in eastern North America in response to the combination of factors embraced by "acid rain." Forests can be destroyed rapidly; they are replaced, but slowly over a century or more. Similarly, as climates change, forests will migrate in time, but the potential exists for the destructive effects to outstrip the rate of repair. The effect of such changes on forests is clear. When respiration outstrips photosynthesis, plants and other organisms cease growth and ultimately die. In the longer term they are replaced by other species adapted to the new conditions, but in the short term of decades, the effect is 73 Deforestation in the state of Rondonia, in the southwest Amazon basin, as photographed by Landsat in August 1978. The image covers a ground area 185 kilometers on a side, and shows roads cut into the forest. The government later gave away lOO-hectare tracts to individuals who cleared the land. widespread mortality of trees without replacement. This mortality results in the decay of some fraction of the large pools of carbon held in forests of these latitudes with the subsequent release of the carbon as carbon dioxide. The amount of carbon available for such release is large, certainly in the range of hundreds of billions of tons (10'^ grams). This carbon is released in addition to the carbon released through combustion of fossil fuels and deforestation. The most reasonable prognostications are that there will be substantial changes in climate in the forested zones over the next years to decades, and that these changes will be but the forerunner of more changes. Under those circumstances the stabilization of climate required for the re- establishment of forests in zones in which forests have been lost will be long in coming and the impoverishment will be widespread and persistent. An Example The problem can best be understood by considering the transition from the deciduous (leaf- bearing) forests of temperate eastern North America to the coniferous (cone-bearing evergreen) forests of more northern and mountainous regions. As the climate warms, the evergreen forests of the north move further northward. At the Southern boundary of the region, at the transition to deciduous forest, there is widespread mortality of coniferous trees and other species associated with them. In the course of a decade, the mortality may spread many miles northward. The deciduous forest does not migrate rapidly, but slowly. The effect is widespread mortality of trees and other species near their limits of distribution: a wave of biotic impoverishment as profound as any change imposed by the glaciation. The effect is not limited to forests, of course. It will reach to all vegetations, especially those of the middle to higher latitudes where the climatic changes are greatest. But the most important change may be in the carbon cycle itself, for the destruction of forests and soils in this way will almost certainly result in a substantial further release of carbon from biotic pools into the atmosphere. The amount of the release is difficult to estimate, but it has the potential of releasing hundreds of billions of tons of carbon over years to decades, depending on the speed of the warming. Such releases are significant in the global balance and will accelerate the warming. It is difficult to envision any set of parallel changes in the biota that would reduce this positive feedback system over the period of years to decades before the most vulnerable pools of carbon are exhausted. How Can We Manage Such Problems? The transitions discussed are thought by some to be under way at the moment, although the evidence is not yet conclusive in the eyes of all. The potential effects, however, of initiating an indefinite period of rapid global climatic change are profound. They include the possibility of a significant positive feedback through biotic effects, which include the destruction of forests over large areas in the middle to higher latitudes, and the beginning of a continuing disruption of agriculture as climates change around the world. The sudden destruction of forests by air pollution, now being experienced in northern and central Europe and in the eastern mountains of North America, is but a sample of the destruction that appears to be in store as the climatic changes anticipated become reality over the next years. The issue is unquestionably one of the most urgent topics for the agenda of the councils of nations. It strikes at the core of the question of the continued habitability of the Earth at the very moment that the human population is passing 5 billion on its unplanned and uncontrolled upward path. It has a potential for disruption of the human enterprise over a few decades that rivals the chaos of war. The issue will force itself onto the agendas of governments. The question is, when will it be addressed, and how effectively. For the moment, the issue seems to be in the hands of a small group of scientists who struggle with very modest budgets, probably no more than $30 million annually worldwide, to test whether present forecasts are indeed as threatening as they appear to be. All the answers available at present appear to confirm their apprehensions. Several steps are appropriate now. First, there is a clear need for additional detail on the carbon cycle. The terrestrial component is large and, at the moment, neglected. Research to define places and rates of deforestation and the response of terrestrial ecosystems to climatic change is appropriate now. It is not under way. While there 74 Clearing in the Brazilian Antj/on Bcis/n. I fry little ot the lumber is vi/v.iged. The (rees are cut, allowed to dry, and then burned. The cleared land often supports crops for two to three years before being abandoned to become unproductive cattle pasture. is now an excellent monitoring system available for recording the changes in atmospheric carbon dioxide, there is no systematic program to monitor the effects of the great forested zones of the Earth on the atmosphere. Such a program would be cheap and extremely valuable. None is even proposed at the moment. The transfer of atmospheric carbon into the oceans, discussed in the articles beginning on page 9, also warrants more intensified study. But the greatest challenge at the moment is to introduce the issue of climate change into the councils of governments — because solutions will require unified action. That action will involve unified policies in the use of fossil fuels and other sources of energy, and global policies in the management of forests, especially the world's remaining tropical forests. Time is short. Action should be taken before climate change causes widespread social disruptions, not after. George M. Woodwell is founder and Director of the Woods Hole Research Center, Woods Hole, Massachusetts. References Buringh, P. 1984. Decline in organic carbon in soils of the world. In The Role of Terrestrial Vegetation in the Global Carbon Cycle: Measurement by Remote Sensing, C. M. Woodwell, ed. pp. 91-109. New York: John Wiley and Sons. Cleveland, W. S., A. F. Freeny, and T. E. Craedel. 1983. The seasonal component of atmospheric CO2: information from new approaches to the decomposition of seasonal time series. /. Geophys. Res., 88: 10934-10946. Houghton, R. A., |. E. Hobbie, ). M. Melillo, B. Moore, B. |. Peterson, C. R. Shaver, C. M. Woodwell. 1983. Changes in the carbon content of terrestrial biota and soils between 1860 and 1980: a new release of CO2 to the atmosphere. Ecolog. Monog. 53: 235-262. )ones, P. D., T. M. L. Wigiey, and P. B. Wright. 1986. Global temperature variations between 1861 and 1984. Nature 322: 430-434. National Academy of Sciences. 1983. Changing Climate. Report of the Carbon dioxide Assessment Committee, NAS-NRC. Washington, D.C.: National Academy Press. Pearman, C. I., and P. Hyson. 1980. Activities of the global biosphere as reflected in atmospheric CO2 records. I. Geophys. Res. 85: 4457-4467. Schlesinger, W. H. 1984. Soil organic matter: a source of atmospheric CO2. In Woodwell, ed. The Role of Terrestrial Vegetation in the Global Carbon Cycle: Measurement by Remote Sensing, pp. 91-109. SCOPE 23. New York: John Wiley and Sons. Strain, B. R., and |. D. Cure, eds. 1985. Direct effects of increasing carbon dioxide on vegetation, 282 pp. U. S. Dept. of Energy, National Technical Information Service, Springfield, Virginia 22161. Woodwell, C. M., and W. Dykeman. 1966. Respiration of a forest measured by CO2 accumulation during temperature inversions. Science 154: 1031-1034. Woodwell, G. M., and R. H. Whittaker. 1968. Primary production in terrestrial ecosystems. Amer. Zoologist 8: 19-30. 75 Spaceborne Observations in Support of Earth Science by D. James Baker^ and W. Stanley Wilson I he space program has had a profound impact on science and technology dating from the launch of the first satellite more than a quarter of a century ago. We have been able to send planetary probes to the near and far planets in our solar system. We have made the first infrared measurements of the deep universe outside the influence of the atmosphere, and are preparing to launch the first Space Telescope. We have developed a highly successful commercial communications industry based on satellite technology. But the satellite technology with perhaps the most far-reaching consequences for life on Earth is that which studies Earth itself. One of the first, and continuing, successes of the space program has been this "remote sensing" of Earth. Weather satellites were among the first to fly: the Television and Infrared Observation Satellite, TIROS, was launched in 1960 to give the first view of Earth from space. Those global cloud pictures from TIROS were the first step toward today's global weather observing system that provides data for modern weather forecasting. Figure 1 illustrates the present global system that depends on five geostationary and six polar-orbiting Earth-looking satellites that continuously monitor atmospheric, terrestrial, and oceanic conditions. The U.S. part of this system is operated by the National Oceanic and Atmospheric Administration (NOAA). Over the years, satellite technology has evolved to give us much more than weather observations. We now can study physical, biological, chemical, and geological processes from space. From information laboriously pieced together from many observations — ground based, and from ships at sea — scientists have learned that environmental change is global in scope and that such processes are all linked. The development of satellite technology and of high-speed computing has come at the same time as this realization of the need to study processes on a global scale. The global view of the Earth that satellites provide has stimulated this new view of the Earth, and of the forces that drive environmental change. For example, we have learned that we need 76 understanding of global processes so we can provide, as much as is possible, global warning of potential major environmental change. This includes natural effects such as the El Nino/Southern Oscillation (see Oceanus, Vol. 27, No. 2), long-term climate change, volcanic eruptions and earthquakes, and those effects where mankind itself may be affecting the environment. The other articles in this issue have shown that increased carbon dioxide, methane, and other gases can lead to global temperature increases; we see increasing "holes" or depletion of ozone in the atmosphere over the Antarctic and Arctic; and we know that deforestation can lead to climate change. Spaceborne measurements can provide much of the data that is required to study these problems. Today, we have satellites that monitor the energy that comes from the sun so that we can correlate changes in this driving force with changes in the Earth's response, which is our climate. Satellites measure the temperature and winds in the atmosphere for input to weather forecasting models. They also provide data on the concentration of chemical constituents in the atmosphere from carbon monoxide to ozone to a variety of other such "greenhouse" gases — those that are effective in causing global warming. Satellites have helped to reveal how the physical state of the ocean, the geology of the land surface, and the structure of marine and terrestrial ecosystems are all critical elements in global environmental change. Following the developments building on TIROS, NASA launched a series of experimental Earth observing satellites called Nimbus. This series, the last of which is Nimbus-7, together with the Seasat dedicated oceanic satellite, provided the first successful measurements of a number of oceanic and atmospheric parameters. Nimbus-7, launched in 1978, is still flying after eight years, it has instruments that monitor both the radiation coming from the Sun and that reflected back to the satellite from the Earth. From the energy content at different spectral bands, it is possible to monitor parameters ranging from the changing heat budget of the Earth, to Operational Earth Observation Satellites METEOSAT (ESA) 0° LONGITUDE GMS (Japan) 140° E Figure 1. The current civilian operational satellite system includes satellites from five different countries and the European Space Agency. The geostationary satellites view the earth from a distance of 22.500 nautical miles; the polar-orbiting satellites are at a distance of about 520 miles. (The recently-launched French SPOT satellite is not shown.) chemical constituents in the atmosphere, to chlorophyll concentrations near the ocean surface. The Nimbus series is perhaps the most successful Earth-looking satellite program ever launched. Seasat, also launched in 1978, demonstrated that global measurements of wind, waves, and the shape of the sea-surface and ice topography could be achieved with an active radar system: short radar pulses sent and received back at the satellite reveal, by their reflected shape and timing, the characteristics of the surface. Satellite technology has come of age at the same time as the development of new computers. The greatly enhanced computer capability available today ranges from personal computers capable of detailed image processing to super-computers that can yield global weather forecasts in minutes. These computers from small to large are being linked by high capacity communication networks. These new computers and networks will allow the science research and operational user community to cope with the data sets from satellites which, because of their global coverage, are much larger than the Earth sciences community has grown accustomed to. In fact, dealing with such data that comes in at large rates will require a new outlook by scientists — instead of spending long periods of time collecting a few data points, in this new age, great amounts of data will be collected in short periods of time. Pattern recognition and other techniques for dealing with large data sets will be essential. It may require a whole new generation of scientists to make this transition. During the same time that the new technology has given us a global view of the Earth, there has been a growing recognition that, in order to understand global environmental change, the Earth must be viewed as an interactive system. The impact of this relatively new scientific viewpoint is amply demonstrated by the exciting science presented by the other articles in this issue. It is clear that the new scientific insights have come from data collected by in-situ measurements as well as satellite measurements, but only satellite measurements can provide the necessary global coverage of the relevant processes. Satellite Plans for 1986-1995 What are the satellite systems that are likely to be available to Earth science to address the many problems described in this issue? The satellite systems and missions, now in place or planned, come from a variety of programs supported by U.S. and non-U. S. federal agencies and industry. Although the systems were not necessarily designed and promoted with global Earth sciences as the 77 Oceanographic satellites: past, present, and future. Seasat was launched June, 1978, and (ailed October, 1978. Nimbus-7 was launched October, 1978, and continues to operate. Geosat was launched in early 1985, and is now flying and producing data. The NOAA/J1ROS satellites are part of a continuing operational series. NROSS is approved for a 1990 launch, and TOPEX is approved for a mid- 1 99 / launch. primary impetus, the total system described will address a significant traction of the overall Earth science picture. In meteorology, with the longest history of satellite-based measurements, an international array of operational satellites is in place and will continue. The array consists of five geostationary weather satellites sponsored by the U.S., the European Space Agency (ESA), India, and the Japanese National Space Development Agency (NASDA) that continuously monitor cloud cover, atmospheric temperature and water content, and the space environment of protons, alpha particles, and electrons from the Sun for operational weather and solar forecasts. Since these satellites are in the equatorial plane of the Earth, they cannot view the polar regions. Therefore, for these operational purposes, there are in addition three polar-orbiting meteorological satellites sponsored by the United States (NOAA's advanced TIROS-N weather satellite) and the Soviet Union that provide coverage of the Earth with full polar capability (see Figure 1). These satellites also carry data collection and transmission systems and search and rescue monitors. In the upper atmosphere, the Nimbus series of satellites, culminating in Nimbus-7, demonstrated 78 Winds Over the Pacific Sea-surface winds, a fundamental driving force for ocean waves and currents, have a major influence on the exchange of heat and molecules between the ocean and the atmosphere. Knowledge of their dynamics is an important component in the understanding of our climate. In the early 1990s, NASA plans to fly the next generation scatterometer (NSCAT) on the Navy Remote Ocean Sensing System (NROSS) satellite and an altimeter on the JOPEX/POSEIDON mission; this capability will allow scientists to collect simultaneous global synoptic wind obserx'ations and sea-level data to determine the ocean circulation. The numbers represent various storm and wind systems. Charts based on September 1978 Seasat data. the feasibility of measuring parameters, including atmospheric pollutants, ocean chlorophyll concentrations, and weather and climate parameters. NASA's new Upper Atmosphere Research Satellite (UARS), approved to fly in 1989, is based on results from Nimbus-7. It will provide a coordinated measurement of major upper atmosphere parameters using remote sensing instruments currently in development. These include two instruments being provided by British and French investigators, to measure trace molecule species, temperature, winds, and radiative energy input from, and losses to, the upper atmosphere. It also will make in-situ measurements to determine the flux of electrons and protons, which can have an effect on the dynamics of the upper atmosphere. The UARS measurements take on special significance in the context of the issues raised in the articles by Andreae (see page 27) and by Moore and Bolin (see page 9) earlier in this issue, where it is shown that chemical constituents in the atmosphere other than CO2 are just as important in warming the atmosphere. For land surface measurements of radiation, vegetation type, soil moisture, and geological form, a series of successful NOAA satellites named Landsat has shown that multispectral measurements can be used with good success. An operational land- measuring system based on these results is now in place. The Landsat series is continuing, and is in the process of being transferred to the private sector. Providing that previously agreed-on U.S. Government subsidies are made available, this data series is expected to continue. A Japanese satellite, the Marine Observation Satellite-1, will conduct observations of the Earth similar to those by Nimbus-7, including sea-surface temperature and atmospheric water vapor for three years beginning in 1987. ESA's ERS-1, to fly in 1990, will carry active radar instruments providing data similar to Seasat together with multispectral sensors. ERS-1 is expected to be the first of an ESA-sponsored series of operational Earth remote sensing satellites, designed to increase understanding of global ocean processes and to promote economic and commercial applications. France also has established an operational land measuring program, Systeme Probatoire d'Observation de la Terre (SPOT), based on the Landsat technology. SPOT provides a variety of spectral information, and is expected to continue indefinitely in an operational mode. The data will be essential to the study of global processes at the land surface. Plans for future SPOT sensors may include measurements at wavelengths sensitive to ocean chlorophyll concentration. The data from such measurements as SPOT and Landsat is very precise. Landsat, for example, can distinguish fields of corn from fields of soybeans by the radiation emitted by the different vegetation. Figure 2 shows the operation of a multispectral Landsat sensor in orbit. The land data will be an integral part of the International Satellite Land Surface Climatology Project, which is designed to provide information about the effect of land surface processes on climate change. Land and ice measurements are also planned from satellites with synthetic aperture radar (SAR), a high resolution technique that maps surface properties and topography. The high resolution of the technique, as fine as 30 meters, leads to enormous data rates. As a consequence, there is a need for ground stations for data collection since storage onboard the satellite is not always feasible. Two missions that will carry SAR are now approved: ESA's ERS-1 which will fly in 1990, and japan's jERS- 1 . Each of these also carries other instruments. For the ocean, the currently proposed missions build on the Seasat and Nimbus missions. The Navy's Remote Ocean Sensing System (NROSS) is an approved mission scheduled to fly in 1990 to provide operational sea state, surface wind, sea- surface temperature, and sea-ice distribution data on improved oceanic nowcasts and forecasts. The instruments will include an active radar, altimeter, scatterometer, and radiometers. ESA's ERS-1, also scheduled to fly in 1990, will have a similar complement of instruments. NROSS and ERS-1 will provide the first continuing global measurements of these parameters which will be as important to research in understanding global ocean processes as they are in operational ocean forecasting. Global ocean currents, a key parameter in global change, can be measured with precise (within a few centimeters) data on ocean surface topography. Such measurements can be made with a satellite altimeter. The joint U.S./French mission TOPEX/POSEIDON, which has been in the design stages for several years, is now being prepared for flight in 1991. TOPEX/POSEIDON will have precise altimeters and sufficiently accurate orbit determination to provide for the first time global measurement of the major currents of the ocean. Together with the surface wind measurements that will come from NROSS and ERS-1, the ocean current measurements from TOPEX/POSEIDON will provide a global data set that will be the centerpiece of an important new scientific program, the World Ocean Circulation Experiment (WOCE), described on page 25. WOCE, which includes a major in-situ set of ocean observations ranging from sea-level measurements to arrays of drifting floats, to global hydrographic, chemical, and nutrient surveys, is aimed at describing and understanding ocean circulation well enough to make coupled models of the atmosphere-ocean system for prediction of climate change. The global-scale wind data will also be used in the Tropical Ocean-Global Atmosphere program (TOGA), which is aimed at understanding and forecasting El Niiio-type events. Ocean productivity has been mapped by the Nimbus-7 Coastal Zone Color Scanner (CZCS). Today we have from the CZCS the first ocean basin- wide maps of ocean color, related to the concentration of chlorophyll and hence productivity and suspended sediment (page 23). The CZCS has provided data during the 8-year lifetime of Nimbus- 7, and designs have been prepared for a follow-on instrument — the Ocean Color Imager. Arrangements for flying such an instrument on Landsat-7 and for including a channel sensitive to ocean color on the SPOT follow-on missions are in the discussion stages 80 Multispectral Scanner (MSS) Sensor MSS Detector Details r Bands -i 1 2 3 4 — , 4 Bands 6 Detectors Per Band Along Track Direction West -^ East Scan Direction Band Spectral Ranges 1 5-6 2 6- 7 3 7- 8 4 8- 11 Ground IFOV All Bands — 83 Meters Data Rate — 1506 Mbps Quantization Levels — 64 Figure 2. Operation of a Landsat-type sensor in flight, showing the swath width and orbit directions. now. In addition, an instrument including more than 100 spectral bands — the Moderate-Resolution Imaging Spectrometer (MODIS) — is currently being designed to fly as part of an Earth Observing System in the mid-1990s. Because of its wide spectral range, the MODIS will provide monitoring of changing land use, deforestation, and climatically induced desertification as well as ocean chlorophyll and suspended sediment. Geodesy and geophysics, including the structure of the gravity and magnetic fields of the Earth, will be addressed first by orbit determination (gravity only) and later by dedicated satellites. A Magnetic Field Explorer (MFE) is under consideration to measure the global scale field. A Geopotential Research Mission (GRM), flying an accurately tracked proof mass and a magnetometer, is also proposed to measure both the gravity field and the magnetic field of the Earth. Together, the MFE and the GRM will provide the necessary global measurements of gravity and magnetic fields. These missions should fly in the mid-1990s. The output of the Sun, as it directly affects the Earth and as it interacts with atmosphere, must also be monitored. An Earth Radiation Budget Experiment (ERBE) has been carried on an experimental basis on Nimbus-7; an Earth Radiation Budget Satellite was launched in 1984 and is still flying, and an operational radiation budget system is on the NOAA Advanced TIROS-N weather satellite. All of these monitor the output of the Sun as well as the changing reflected radiation from Earth. An International Solar-Terrestrial Program/Global Geospace Experiment is currently approved to help address these issues starting in 1989. Immediate Opportunities, Challenges What can we do to be ready to exploit the full potential of these opportunities and challenges? The science community through a number of international programs, past, current, and planned, has developed an approach to carrying out large interdisciplinary programs that involve both satellites and in-situ instrumentation. These range from the International Geophysical Year (IGY) in 1957 to the World Climate Program, now under way, which includes the Tropical Ocean/Global Atmosphere and World Ocean Circulation Experiments discussed previously. Planning for future programs builds on these, and is evolving towards an International Geosphere-Biosphere Program (IGBP) designed to understand global environmental change on a broad scientific front. The IGBP will focus on interactions between physical, chemical, biological, and geological processes, and on global biogeochemical cycles. This will allow, for the first time, a full study 81 Satellites provide a way to observe phytoplankton blooms, such as shown here surrounding the Pacific coast ot Baia California. The phytoplankton abundance, measured as chlorophyll pigmentation concentration, was calculated from ocean color data collected by the Coastal Zone Color Scanner (CZC5) on NASA's Nimbus-7 satellite: (I) indicates high-phytoplankton concentration caused by wind-driven upwelling; (2) shows plankton-rich water mixing with the California Current; (3) is the Costa Rica Current flowing north bnnging warm water to meet the cooler southward-flowing California Current; (4) indicates low-nutrient surface water flowing from the south, resulting in low levels of phytoplankton; and (5) is a shallow basin where strong tidal current cause intense mixing, resulting in high-phytoplankton abundance. An improved version of the CZCS, the Ocean Color Imager (OCI) has been proposed for the early 1990s. Beyond that, a new instrument, the Moderate Resolution Imaging Spectroradiometer (MODIS) is planned for the mid-1990s. These instruments are needed to further our understanding of the magnitude and long-term variability of regional and global ocean productivity. 82 Sea-Surface Topography and Ocean Bathymetry Both the sea surface and the seafloor have a topography. The upper figure shows sea level obtained from an altimeter aboard Seasat in 1978. Blues indicate surface lows, and yellows indicate highs. Of the numbered features shown, the most surprising is an unusual pattern of low-altitude bumps (7) in the sea level. These bumps may indicate the location of upwelling plumes of molten rock within the interior of the Earth. The TOPTX/POSEIDON mission, planned for 1991, will provide still more accurate sea level data. The lower figure shows ship soundings assembled into a computer image of seafloor bathymetry. Blues indicate deep, and yellows indicate shallow areas. of the cycles of such life-sustaining elements as carbon, sulfur, phosphorus, nitrogen, and oxygen. These new programs offer a marvelous opportunity to document and understand the Earth. The satellite missions required are either approved or in place. But how well will these missions meet 83 Table 1. The EOS Earth Observing System (Instruments). Instrument Measurement Spatial Resolution Coverage 1. Automated Data Data and command relay and location of Location to 1 km for buoys. global, twice daily Collection & Location remotely sited measurement devices to 1 m for ice sheet System (ADCLS) packages SISP — Surface imaging & Sounding Paclcage 2. Moderate Resolution Surface and Cloud imaging in the visible and 1 km X 1 km pixels (4 km x global, every 2 days during imaging Spectrometer infrared .4 nm-2.2 nm, 3-5 urn, 8-14 ^m 4 km open ocean) daytime plus IR nightime (MODIS) resolution varying from 10 nm to .5 )im. 3. High Resolution Surface imaging .4-2.2 nm. 10-20 nm spectral 30 m X 30 m pixels pointable to specific Imaging Spectrometer resolution targets, 50 km swath (HIRIS) width 4. High Resolution 1-94 GHz passive microwave images in 1 km at 36.5 GHz global, every 2 days Multifrequency several bands Microwave Radiometer (HMMR) 5. Lidar Atmospheric Visible and near infrared laser backscattering vertical resolution of 1 km. global, daily atmospheric Sounder and Altimeter to measure atmospheric water vapor, surface topography to sounding; continental (lASA) surface topography, atmospheric scattering 3 m vertical resolution topography total in 5 properties every 3 km over land years SAM — Sensing with Active Microwaves 6. Synthetic Aperture L, C, and X-Band Radar images of land, ocean. 30 m X 30 m pixels 200 km swath width daily Radar (SAR) and ice surfaces at multiple incidence angles coverage in regions of shifting sea ice 7. Radar Altimeter Surface topography of oceans and ice. 10 cm in elevation over global with precisely significant wave height oceans repeating ground tracks every 10 days 8. Scatterometer Sea surface wind stress to 1 m/s, 10° in one sample at least every 50 global, every 2 days direction Ku band radar km APACM — Atmospheric Physical & Chemical Monitor 9. Doppler Lidar Tropospheric winds to 1 m/s doppler shift in 1 km vertical, 2° longitude. global, twice daily surface laser backscatter 2° latitude to 100 mb 10. Upper Atmosphere Upper atmospheric winds to 5 m/s, doppler 3 km vertical, 2° longitude. global, daily Wind Interferometers shift in O2 thermal emissions 2° latitude 11. Tropospheric Trace chemical constituents of the varies from total column global, daily, surface to 100 Composition Monitors troposphere density to 1 km vertical, from r to .1° horizontal mb 12. Upper Atmosphere Trace chemical composition passive emission 3 km vertical 2° longitude. tropopause to 1 20 km Composition Monitors detectors at wavelengths from UV to 2° latitude global daily day and microwave night coverage 13. Energy and Particle Solar Emissions from 150-400 nm, 1 nm total solar output roughly continuous Monitors spectral resolution. Earth radiation budget sampling, at least twice Total Solar irradiance daily for solar Particles & fields environment observations the scientific program needs? There are several issues that must be addressed. The satellite-based measurements must be adequately calibrated and validated: calibrated to ensure that the sensor operates properly and that instrument drift is low or understood, and validated to ensure that the quantity measured can be unambiguously tied to a geophysical parameter. For example, the drift problem can be a major issue — such questions as to whether the Landsat sensors show desertification or just instrument drift are complex, and involve both calibration and validation. Data simultaneity is also an issue — it is essential that physical, chemical, biological, and geological processes important to global change be measured at the same time. This puts constraints on mission timing, since satellites need to be scheduled several years in advance. Current planning, as is clear from the list of instruments planned for the Earth Observing System (EOS), shown in Table 1, is leading towards a large set of sensors in orbit. together with in-situ field programs to be operating by the end of this decade. Synopticity is also an issue — processes must be globally monitored over periods short compared to significant change. This means that missions must range from rapid (daily) sampling for atmospheric, land surface, and ocean processes, to decadal sampling for the Earth's magnetic field. Current plans, provided they are carried out, are consistent with this need. All of these issues are related to data continuity. The previous articles in this issue have shown that long-term (at least 20 years) data sets will be required to document and understand global change. How can we actually realize the necessary long-term observations? We need to ensure overlap of calibration as satellites and instruments are replaced, we need to provide for a continuing series of in-situ validation studies, and we need to make sure that appropriate data is saved and archived. 84 Archival and distribution of data will be an increasingly difficult issue as the new satellite systems provide large quantities of data. Fortunately, advances in computers and communication technology are also rapidly providing a means for handling these large quantities. For example, all the data from the TOPEX/POSEIDON mission, sea- surface topography every 150 kilometers collected every 10 days for three years, amounts to about 3 x 10'' bits of data which would fit on just 3 optical disks. It is important that the various archives speak the same language, so that researchers can integrate and overlay data sets of various types and quality as they analyze data. It will be essential that networks have compatible protocols, so that it is easy to access archives at multiple locations. Even if satellites fly, however, data access is not guaranteed. Both Landsat and SPOT operations raise an issue important to researchers — the need for access to data that is only available through commercial channels at a market price. Since the data cost will be up front, arrangements for providing data for research purposes must be found in addition to normal research support costs. Without such support, the (legitimate) costs of the data which are now being provided for by the government would have the effect of reducing total research support and thus constraining the research that must be done for these techniques to reach their full potential. International agreements also are required for data exchange between U.S. and non-U. S. missions. U.S. agencies must provide support for data storage and distribution. With commercial and military systems, those cost and national security aspects which might preclude access must be addressed both with adequate funding and interagency agreements. Looking to the Next Century The operational and research missions planned for the coming decade — through the mid-1990s — will lay the basis for a new global view of processes on Earth. In anticipation of this view, and in recognition of the need for continuous long-term measurements, plans are being made now for the implementation of a series of missions tailored specifically for earth science. NASA's Earth Observing System (EOS) is designed to meet both the short term and long term needs of the program. EOS, which includes a full set of instrumentation for observing land, atmosphere, ocean, and ice from polar-orbiting satellite platforms, is planned as part of the space station. The capability of the shuttle for on-orbit servicing and repair is an important factor for achieving long-term continuity of measurement in the EOS plan. The co-orbiting platform of the space station, which will view tropical and sub-tropical regions, could be used, for example, for tropical rainfall monitoring, a key parameter in climate. There also will have to be an array of in-situ stations — a recent National Academy of Sciences report has recommended a Permanent Large Array of Terrestrial Observations (PLATO) for this purpose. PLATO would provide the long-term calibration and validation data that is so essential. FHow is such a long-term plan to be kept on track? The science community needs to develop, refine, and prioritize science plans based on results from on-going programs. The Federal agencies need to agree, with appropriate input from the science and operational community, on the relative roles of NASA, NOAA, the National Science Foundation, the Department of Defense, the U.S. Geological Survey, and other interested agencies. It is clear that, for example, NASA will play a key and central role in satellite technology and associated research and development; NOAA in operations, associated mission-related research, and data management; and the National Science Foundation in basic research. Together, these groups must work with the international and commercial communities to ensure adequate coordination. With the new plans, proper coordination, and the emerging new global science results, it should be possible to obtain consensus and the financial support from all the parties involved. The feasibility has been shown and the opportunity is there — we must move now to achieve a major new program for the 1990s. D. lames Baker is President of joint Oceanographic Institutions, Inc., and chairs the Committee on Earth Science of the Space Science Board of the National Research Council. W. Stanley Wilson is Chief of the Oceans Branch of the National Aeronautics and Space Administration, a post he has held since 1979. Additional Reading Joint Oceanographic Institutions Incorporated. 1984. Oceanography From Space: A Research Strategy for the Decade 1985-1995. 53 pp. Washington, D.C.: |OI. McElroy, ). H., J. Clapp, and |. C. Hock. 1986. Earth Observations: Technology, Economics, and International Cooperation. Presented at the NAE/RFF Symposium on Explorations in Space Policy, lune 1986. Washington, D.C.: National Academy of Engineering. NASA and NOAA. 1986. Space-Based Remote Sensing of the Earth and Its Atmosphere: A Report to the Congress. 126 pp. Washington, D.C.: NASA. National Aeronautics and Space Administration. 1986 (in press). Earth System Science: A Closer View — Detailed Scientific Rationale. Washington, D.C.: NASA. National Aeronautics and Space Administration. 1986 (in press). Program and plans for FY 1986/87: Earth Science and Application Division. Washington, D.C.: NASA. National Commission on Space. 1986. Pioneering the Space Frontier. 211 pp. New York: Bantam Books. Waldrop, M. M. 1986. Washington embraces global earth science. Science 231: 1040. 85 Robert A. Frosch Unafraid to Take Risks V/ui "UnfTap lietly competent." ip(3able." "Combines two great abilities: a great technical ability with an unusual ability to understand and analyze a problem." "Combines the qualities of a great administrator with those of a great scientist." by James H. W. Hain Who are we talking about? Clearly, we are speaking of an unusual individual. Robert A. Frosch has administered or managed a variety of the largest research and development programs on the national and international level, and operated within the higher levels of government and research institutions. He has been Director of the Hudson Laboratories, Deputy Director of the Defense Advanced Research Projects Agency (DARPA), 86 Assistant Secretary of the Navy tor Research and Development, Assistant Executive Director of the United Nations Environment Program, Associate Director of Applied Oceanography at the Woods Hole Oceanographic Institution (WHOI), and Administrator of the National Aeronautics and Space Agency (NASA). Today he is Vice President of General Motors in charge of the Research Laboratories. Along the way, he has been involved with some fascinating and important projects. In learning something about the man, we also are afforded glimpses into a few of the pivotal chapters from the history of science and technology. Hudson Labs In the late 1940s, Maurice Ewing, j. L. Worzel, H. B. Sherry, and others did some experiments at Bermuda on the detection of snorkeling submarines by acoustic means. From this work, and subsequent work of Ewing and others, the field of underwater acoustics took a large step forward — when the long-range propagation and low attenuation of low-frequency acoustic waves in the ocean were discovered. Both Bell Laboratories and the U.S. Navy became interested in the applied aspects of this discovery. In particular, there was great interest in the detection of submarines at long distance by means of these low-frequency waves. Isadore I. Rabi, Nobel Prize winner, and a physicist at Columbia University, was an advisor to the Navy at the time. As a result, the Hudson Laboratories were established in 1951 at Dobbs Ferry, 30 miles up the Hudson River from New York. Eugene T. Booth was the first director, and Bob Frosch came aboard as a house theoretician just prior to finishing his doctorate in Physics at Columbia. Basically, Hudson Labs was a classified research laboratory, affiliated with Columbia University, supported by the Office of Naval Research, and geared to the Navy's acoustic antisubmarine-warfare The Hudson Laboratories, Dobbs Ferry, New York. A navy contract laboratory affiliated with Columbia University, they opened in 195 1 and closed in 1969. At their peak, the laboratories employed 350 persons. (Photo courtesy of Columbia University) program. There was concern that the field had been changing, and that if something like World War II were to reoccur, the United States would be unable to cope with the submarine problem. Frosch recalls, "We did a lot of work in ambient noise, and sound propagation. My specialized interest was trying to make some theory of the ocean that would enable us to predict long-range sound propagation — over hundreds of miles. The gospel at the time was that all sound was incoherent in the ocean. Also, one of the problems was that the professional acoustics people were terribly impressed by the variability of the ocean — partly because they tended to work at a few hundred to a few thousand cycles. "We started to work at a few tens or a couple of hundred cycles. We couldn't see why the ocean was going to be so damn variable, and we couldn't see how motions could take place fast enough over those scales. Our conclusion was that, in fact, it was probably pretty predictable. We were able to demonstrate that, and also that you could measure phase over hundreds of miles — in the 30-, 50-, and 100-cycle range. We were able to measure well enough so that one could use Doppler Shift to tell which way a sound source was traveling. As a result, one could locate things quite accurately, and get a reasonable picture of what was taking place. "Like other researchers in the field, most of the work we did was with artificial sources — with explosions and noise makers. It was easier to do the experiments with controlled sources. We did a lot of work down in the Puerto Rico Trench. Because of its depth, it was a simple case. Most of the work, of course, was classified, but our mapping of the seamounts, a by-produ( t of the acoustics work, did show up in the literature. "We were looking for seamount situations that would provide certain types of screening. We found that the New England seamount chain had never really been surveyed. We also found several that had never been charted before. Even though we had a lot of fun doing the work, we got into a 87 In the course of acoustic experiments, tiie Bermuda-New England Seamount Arc was mapped. Charting of the 24 seamounts was done through a joint effort by Hudson Laboratories, the Woods Hole Oceanographic Institution, and the Navy Hydrographic Office. (After /. Northrop, R. A. Frosch, and R. Frasetto. 1962. Ceol. Soc. of America Bull. 73: 587-594) mammoth battle with the hydrographic office — because we didn't know how you were supposed to survey a seamount. We had looked at the geometry and invented the "correct" way, which didn't happen to be the "book solution" that they used — so there was a lot of fuss when we sent in the data. Somewhere, we have a letter explaining how we did it all wrong." One activity of the times was the placing of hydrophones on the seafloor for listening — primarily to submarines. Typically, these were tied in to shore installations via cable. The placement of these hydrophones required detailed knowledge of how the propagation of sound from the deep ocean to sites on the continental shelf was affected by ocean conditions and bottom shape. This involved some complex mathematics. In pre-computer days, arriving at the results of extended mathematical computations took on a whole different flavor. Frosch recalls working on a rather elaborate ray tracing for acoustic propagation off the east coast of the United States: "Because the sound velocity change was very variable, we had to do an inch- by-inch computation. So, we constructed a computer — in pre- computer days. We hired eight or nine young people, set up a flow program, and provided each of them with one of those old electromechanical calculators. We set up Snells Law, a difference equation system, and a flow sheet: 'You do these.' 'You get this pile of numbers.' 'You do this computation, and pass it to the next one.' 'At the end, you put a point on the graph.' They spent every day, eight hours a day, eight or ten of them, tracing rays, for six or seven months, in order to make one ray tracing." In the course of his stay at Hudson Labs (1951 to 1963), Frosch worked on the acoustic exploration of the Norwegian Sea, set up hydrophone arrays, continued acoustic studies of the ocean, and worked on signal processing. He became Associate Director, and then Director, of the lab. At that point, the lab had increased to a staff of between 250 and 300, with an annual budget of a few million dollars — not a tiny operation. Near the end of his tenure at the Hudson Labs, an unexpected test came — not only for Hudson Labs — but for much of the community doing the undersea research of the day. America experienced her first nuclear submarine accident when the 3,750-ton nuclear submarine Thresher sank during trials 220 miles east of Boston on April 10, 1963. This event tested, and indicated the difficulties and shortcomings of the still-young field of offshore navigation and sea-bottom searches. Frosch, then director of Hudson Labs, was at a meeting of the Undersea Warfare Research and Development Planning Council in Washington, D.C. Frosch recalls how he learned of the accident: "We were having a dinner, at which several of the senior Navy flag officers were present. From time to time, an aide would come in and whisper in an Admiral's ear, go away, and reappear looking quite concerned. Towards the end of the dinner, one of the senior flags came back and told us what the problem was. And, that among other problems, the Navy really didn't have the capability to search for a submarine that had gone down in that depth of water. We sat at dinner and organized the search, and everybody went home. The next morning we put together a fleet of search ships — whatever we had at FHudson Labs, Woods Hole, Lamont-Doherty, and the Navy — and went out and started looking. We used as search equipment anything that anybody had been experimenting with — including towed cameras from Woods Hole, various sonars, and some Navy equipment. The Navy's advisory group essentially became the search group. "We developed our own protocol and our own mapping systems, and just went at it. "The sonar situation was almost hopeless, because it was close in on the continental slope and very rough terrain. From the echoes on these experimental sonars, it was difficult to know what you were looking at. Everything looked like a submarine — we were having a terrible time. "Compared to present deep-sea surveys, our equipment was primitive. The towed camera was essentially a 35-mm camera with extra long rolls of film. You did your tow; you pulled it up; processed the film; and then hoped you could find where it was you thought the camera had been when the photos were taken. Not only was the surface-ship navigation sketchy, the underwater- navigation system was also crude. We had the beginnings, but in reality you only kind of estimated where the camera, towed a couple of miles behind you, had been. We had an interesting time. "Finally, the search group found it, with a deep-towed Woods Hole camera as I recall. We kind of stumbled onto the edge of the debris field." On or about June 24, 1963, the wreck was finally pinpointed. The submersible Trieste then surveyed and photographed the remains of the Thresher, lying at a depth of 8,400 feet. DARPA Science only rarely exists in the fabled rarified atmosphere of the ivory tower. In truth, many of the important scientific advances have come from man's baser instincts — often related to military and defense interests. According to Allan Cox in his 1973 book, "Plate Tectonics and Geomagnetic Reversals," such was the case with several of the research lines leading to the synthesis of plate tectonics theory. Among the research central to the synthesis was the precise plotting of earthquake epicenters — a plotting which provided some of the most precise information available about the location of plate boundaries. This plotting was in fact a by-product of defense interests. in February, 1958, the U.S. Government created the Defense Advanced Research Projects Agency (DARPA). Like a great deal of other U.S. activity of the time, its creation was, in part, a reaction to the launching of Sputnik. DARPA was formed as a central agency for technology R&D for the Department of Defense. The agency represents, at least in part, the high-tech military, and shows how defense needs often guide decisions to pursue new technologies. In 1963, Frosch and his family moved to Washington, D.C. This was just after the limited test ban treaty had been signed that covered nuclear testing underwater, in the atmosphere, and in space. Frosch became Director for Nuclear Test Detection (Project VELA) within DARPA. He was responsible for the national program to insure that the U.S. had systems to determine whether anyone had violated the treaty, and determining whether The LISS Allegheny, one of tour research vessels assigned to the Hudson Laboratories. It was from this vessel that much of the early Hudson Labs underwater acoustics research and mapping was conducted. (Photo courtesy of Columbia University) 89 enough confidence could be achieved in a detection system to do an underground seismic treaty. He also continued seismic-related research. Frosch recalls: "The underwater part was the easy part, because a nuclear device, even the smallest you can think of, creates such a large explosion that you have no trouble detecting it. The space aspect was being covered with the VELA satellites that we launched. There was also a system for atmospheric detection. But, a lot of our effort was spent on the seismic detection. It was interesting that I adopted some of the techniques we had been using in underwater sounds into seismology — building large arrays and other devices. I worked very closely with Frank Press for a while. I like to say that Frank and I dragged seismology into the 20th Century as an instrumental science. In fact, in many ways, we did just that. The field never had any money or attention. We gave it a lot of both. There was a point when the principal financial support of nearly every seismologist in the United States was linked to the DARPA Nuclear Test Detection Program. We had contracts with everybody. Not only in the U.S., but with people around the world. It was in about 1966 that we finished up the process of building up the World Wide Standardized Seismographic Network." Navy R&D In the spring of 1966, Frosch moved onto his next plateau. He became the Assistant Secretary of the Navy for R&D, a position he held for ^Vi years — the longest tenure in the position since the time when it had been held by Franklin D. Roosevelt. His tenure extended from the second half of the Johnson administration through the first Nixon administration. Frosch recalls, "I didn't know anything about airplanes. I remember that within days after I i^mm- WOODS HOLE Marine ariisi VVesi Fraser has captured Woods Hole on an early summer evening in this signed limited edition print. i- YANKEE ACCENT^ 11 Wianno A\cnuc' Osierville, Massathusctis 02655 617 428-2352 had been sworn in, Paul Nitze (Secretary of the Navy) called me in and said, 'Do you know anything about airplanes and airplane engines?' When I said, 'Not really,' he replied, 'Well, you are going on vacation. Take along a couple of books on airplanes and turbine engines. There is this thing called the F1 1 1-B, and 1 want you to take responsibility.' "So, I did what any scientist would do. 1 talked to a couple of people, got a couple of books, and sat on the beach and read about turbine engines." Admiral Tom Connelly, Chief of Naval Operations at the time, recalled in a recent public television documentary, "Wings Over Water: Naval Aviation and Foreign Policy," that when it came time to testify before the Senate Armed Services Committee on the Navy's acquisition of the airplane, "Frosch knew it was no good, and I knew it was no good." The Navy didn't buy any. Then, or ever. Robert W. Morse, Scientist Emeritus at the Woods Hole Oceanographic Institution, and Frosch's predecessor as Assistant Secretary of the Navy, provides a further insight into the Navy job: "The Navy has a large R&D program. There is also a dynamic within the Pentagon, and a huge bureaucracy. The Assistant Secretary has to deal with the admirals, legislative people, and the Secretary of Defense's office, which tends to be civilian technical people. He has the job of criticizing projects and programs internally, and defending them externally. He also has the responsibility for presenting the budget to the Congress, and for interpreting it to the Secretary. judging by Frosch's longevity in the position, it is probably safe to say that he balanced these requirements effectively. Kenya In 1931, after selling her 6,000- acre coffee plantation, Baroness Karen Blixen left Africa. Her home on the outskirts of Nairobi, Kenya, became the setting for 90 Out of Africa, written under her penname, Isak Dinesen. Frosch, his wife Jessica, and his two daughters, Elizabeth and Margery, arrived in Kenya 42 years later (in 1973), some time after high-rises had come to Nairobi. He held the position of Assistant Executive Director of the United Nations Environment Programme. It was here perhaps, that Frosch met his match. "While it was interesting to look at the world from the global environmental point of view, and to be in the UN, it was frustrating. Even though I esteemed myself with being a reasonably adept bureaucrat by then, I found it difficult not to run out of patience with the bureaucratics of the organization. "The UN is frustrating because it is bureaucratically very complex, and politically very complex. The method of doing things was slow and circumscribed. There was negotiating just to see if you could get somebody to agree to start to begin to get ready to collect money to do something." After two and a half years in Africa, Frosch accepted an appointment as Associate Director for Applied Oceanography at the Woods Hole Oceanographic Institution. His stay there was abbreviated by a call from the President. NASA The Apollo-Soyuz Test Project in July of 1975 ended America's era of expendable manned spacecraft. Mercury, Gemini, Apollo, and Skylab were history. Six years later, American astronauts returned to space with the revolutionary space shuttle. Bob Frosch was in Geneva when the call came. Frank Press, the President's Science Advisor, was on the line, "The President would be interested in talking with you. Can you be here tomorrow?" On June 21, 1977, Frosch was sworn in as the fifth Administrator of NASA. The 3^/2 year tenure of Frosch at NASA was highlighted by a diversity of satellite activity, and the continuing development of the space shuttle. Development of the shuttle meant advancement in new technological frontiers. Among the items were the three hydrogen-fueled main engines of the shuttle. As part of the orbiter, they had to be built for repeated use. And, they had to be throttleable. Moreover, their specific impulse had to be higher than any yet made. Another principal item was the array of more than 30,000 individual tiles that replaced the typical heat shields used in Mercury, Gemini, and Apollo. The tiles were required to last from flight to flight, and not be burned away. They also had to be flexible enough to avoid cracking, yet bond securely to the metal of the orbiter. The material in the tiles was such that they could be red hot on one side, and cool enough to touch with one's bare hand on the other. Perfecting these and countless other items took time and effort. In 1977, development was well underway, and a cautious test program was begun. This involved eight "captive" flights and five "free" flights of the Enterprise. In both, the shuttle was carried aloft to an altitude of over 20,000 feet on the back of a 747 aircraft. When performance, stability, control, and safety were satisfactory, the free flights, or "drop-and-glide" tests were begun. These crucial approach and landing tests were conducted at Edwards Air Force Base in California. All development focused toward the first launch, scheduled for sometime in 1981. It was during Frosch's time, too, that the shuttles were named, and it can be guessed that his ocean science background had no small influence: "What happened was that before NASA had a chance to name the Enterprise, the Enterprise got named by the Star Trek groupies. There was an uneasy feeling that unless we made a name rule of some kind, each name would be at the mercy of whatever was in vogue at the time. So we consciously decided to have a rule for naming the shuttles. And, we For moored, fixed position, or prof i measurement of temperature and salinity at depths to 6800 meters: SEA-BIRD'S new SEACAT and SEACAT PROFILER offer proven Sea-Bird conductivity and temperature sensors in a self-contained solid state logging package. These compact instruments provide very high resolution and accuracy, flexible acquisition routines, precision time- bases, and convenient electrical read- out. Pressure measurement available in SEACAT, standard in SEACAT PROFILER. The best sensors cire even better Manufacturers of cable telemetering and internal recording CTl) sv'slems. i<)nductrvnt>'. temperature, dissolved oxygen, pH. and other environmental xiiriafiles // Sea-Bird Electronics, Inc. 1405-132nd Ave. NE, Bel levue. WA 98005 USA. Telephone: (206) 462-8212. Telex: 292915 SBEI UR. 91 The NASA space shuttle Enterprise mated atop its 747 carrier jet during test flights of the arbiter over the Mojave desert of Southern California in 1977. Eight "captive flights" preceded the five "free flight" test glides at Edwards AFB. (Photo courtesy of NASA) looked around, and said, 'since it is intended for exploration, how about great exploring ships? We'll get a list.' And so, we produced a list awfully fast, as I recall — a long list of great ocean- exploring ships. We chose off that list. I guess in the end, I picked, off the list, the names for the first five."* This was also a period of high satellite activity. In the interplanetary satellite category. Pioneer II swept by the icy Saturn on September 1, 1979; Viking made the world's most extensive study of Mars; Voyager photographed Jupiter and Saturn; and the Pioneer-Venus mission, launched in 1978, mapped Venus. Among the applications satellites launched during the period were Landsat 3 (March 5, 1978), several GOES (Geostationary Operational Environmental Satellite) satellites, and Nimbus-7 (October 24, 1978), a weather satellite. Another satellite, Seasat, launched June 26, 1978, produced a wealth of ocean data before its service was cut short 105 days after launch by a power failure. As NASA administrator, Frosch had direct input to fairly major decisions on a rather routine basis. Frank T. Press, President of the National Academy of Sciences, and the President's Science Advisor at the time, recalls, "One time when Frosch came before President Carter, for budget- related discussions, there was an either/or decision. Should we * The space shuttles — Enterprise, Columbia, Challenger, Discovery, and Atlantis — are named for ocean-going ships of discovery. The prototype shuttle Enterprise was the name of a ship used to explore the Arctic in the 1850s (there is some question, however, whether in fact this is the vessel after which the orbiter was named). The Columbia was the ship after which the Columbia River was named, and also the name of the first U.S. Navy vessel to circumnavigate the globe. Challenger takes its name from the British oceanographic vessel that spent nearly four years circumnavigating the globe and doing oceanographic sampling in the 1870s. Ships named Discovery have been numerous. One explored Canada's Hudson Bay and searched for the Northwest Passage in 1610- 1611. Another discovered the Hawaiian Islands and explored the coast of Alaska and western Canada. Two British ships with the name worked in the Antarctic early in the 20th century. Atlantis, research vessel of the Woods Hole Oceanographic Institution from 1931 to 1966, sailed nearly a half million miles on nearly 300 scientific voyages. 92 build a gamma ray observatory satellite, or, should we send a mission to Halley's Comet? Frosch recommended the gamma ray observatory, and that's the one that got decided on." On Being a Manager Today, Bob Frosch is Vice- President, in charge of Research Laboratories at General Motors — the world's largest corporation. His work involves engineering, physics, computer science, mathematics, chemical engineering, biomedical research, and social sciences. He serves on a number of committees to the National Academy of Engineering, WHOI, and others. How is it that Frosch has succeeded in the way that he has? David A. Ross, Senior Scientist, and Chairman of the Geology and Geophysics Department at WHOI, observes: "I can't think of a time when I was at a meeting, when, after Bob Frosch made his statement — which was crisp, without an extra word, and to the point — that I wasn't kicking myself because I hadn't been clever enough to have thought of it myself. At meetings (and this is one of my favorite descriptions), he doesn't require much 'transmission time.' But, he is rarely the first one to say anything. He first lets the conversation go — until he sees what the important elements are. "There is another thing I've always liked. Bob is unafraid to take chances. This has led to some exciting things. "There is one other element to his style. He recognizes other people's talents and abilities, and then gets out of the way. This trait is often easy to talk about; but to actually do it is another thing." Frosch had previously alluded to the latter, but phrased it a bit differently: "I let other people do all the important work." Frosch also quietly offers: "It is useful to know something about the subject. Often, people come into a job not knowing much about it. That is alright if they then proceed to learn about it. But, what always worries me is when a person in this situation thinks that because they have learned management, they don't actually have to know anything about the subject. There seems to be a trend of that kind in business and management schools. I actually got an invitation to address a school on the topic of, 'Do you have to be a technologist to manage technology.' Basically, I thought it was a dumb question. The idea that someone with two years of management education would step in to run a laboratory is bizarre. How it has ever become something that is even quasi- respectable baffles me. "The other practice I learned early on is to assume very little. For example, when I went to NASA, I knew a great deal of engineering, chemistry, and physics. But, working from that base, I quickly did my homework in aeronautics and space technology. One can't come rushing in with the attitude that, 'this is just like the other things I've done so I'm going to proceed right ahead and organize this one too.'" Perhaps Morse said it best. "The name of the game in the technical world is knowledge; you've got to know what you're talking about." Clearly, Robert A. Frosch is one who does. lames H. W. Hain is Assistant Editor of Oceanus, published by the Woods Hole Oceanographic Institution. 93 The Book of Falmouth — A Tricentennial Celebration: 1686-1986, Mary Lou Smith, ed. 1986. Published by the Falmouth Historical Commission. 582 pp. $35.00. Anyone who has enjoyed a stay in Falmouth should have this book. It is a perfect item for those thumbing through times— a rainy afternoon, a quiet night. Celebrating Falmouth's 300th Birthday, the reader is presented with numerous brief historical sketches, articles, and pictures that capture the mood of a town aged by salt spray and sprung from the loins of several "villages." These neighborhoods — Davisville, East Falmouth, Hatchville, North Falmouth, Quissett, Teaticket, Waquoit, West Falmouth, and Woods FHole — form the structure of a community that grew from less than 100 people in 1686 to more than 25,000 today on a year-round basis (65,000 in summer). Bruce Chalmers in an excellent introductory "retrospect" takes the reader from present day Falmouth back to 1602 when the town's "first visiting yachtsman," Bartholomew Cosnold, stepped ashore. In 1660 Jonathan Hatch and Isaac Robinson purchased land from the Indians and cultivated a plantation known as Succannessett which grew and prospered, eventually becoming the Town of Succannessett in 1686. A few years later, the name of the town was changed to Falmouth, "after the port at the mouth of the River Fal in England," where the explorer Cosnold set sail for the new world. The book has been richly enhanced by the liberal use of excellent historic photographs and well designed by Diane Jaroch, Design Manager at MIT Press. Each "village" section, such as Woods Hole, begins with a historical preface, followed by individual reminisces — Val Worthington's Fishing in Woods Hole, jan Hahn's Summer Days at Woods Hole, and Susie Steinbach's delightful Abutters and Lovers. Tucked in the middle of sections one finds Photo Essays — in the case of Woods Hole, one on Steamships. A poem or two is placed like a scented punctuation mark between the passages of the past. One I particularly liked: Putting Away The Boat by Mary Swope The mast's dead weight across our shoulders we gather in the shrouds the trailing stays and with slow steps where poison ivy glows and beach plums darken we form a cortege for summer, turn backs to the beach to shoulder winter's burden. The sadness we hear in the cry of the Canada geese as they vee South is our own, not theirs. hliuJhy .Mary Lo «*»■••-- -.. - - .-w- ■*>; '^ For anyone interested in pursuing the history of Falmouth on the Cape further there is both a Selected Bibliography, pertaining to Cape Cod and the Islands, the town of Falmouth, and the Villages of Falmouth, and an individual Bibliography for specific articles in the book. There also is a listing by William M. Dunkle of historical maps, charts, atlases, architectural and engineering plans, microfilm copies, and aerial photographs of Falmouth. This volume would make a wonderful gift for anyone who harbors fond memories of Falmouth past. Paul R. Ryan, Editor, Oceanus Tricks of the Trade for Divers, by John Malatich and Wayne Tucker. 1986. Cornell Maritime Press, Centreville, MD. 244 pp. + viii. $22.50 The Diver's Reference Dictionary. 1986. Best Publishing Co., San Pedro, CA. 131 pp. $17.50. Tricks of the Trade for Divers is "a commercial diver's book for and by commercial divers," as is succinctly stated in the preface. In recent years, the explosive growth of the offshore oil industry and the resultant development of deep "high-tech" diving techniques have overshadowed the traditional inshore commercial diving branch of the industry. However, many divers make their living performing construction, salvage, and ship 94 maintenance tasks in shallow, usually turbid coastal and inland waters throughout the world. This book is apparently intended to serve primarily the latter class of underwater workers. It is not a basic text in underwater work techniques. The authors assume the reader to have already acquired commercial diving training and/or at least some practical experience. The tone of this no-nonsense book is set from page 1: "During the summer months, every kid with a scuba tank is a 'commercial diver' and will try anything for a ten dollar bill. It is unfortunate that a real professional must sometimes contend with that kind of competition. A man who truly earns his living from commercial diving will be the last one called to do a job where all the amateurs have tried and failed." The authors have shared many hard-won facts and observations from their long careers to give the professional the edge necessary to get the job done. The style is factual, providing practical solutions to underwater work problems in a variety of skills, and is interspersed with informative anecdotes from the authors' experience. The authors are well qualified to dispense this advice. For example, John Malatich began his diving career in 1934. Among many other exploits, he participated in the celebrated salvage of the former French passenger liner "Normandie" (later rechristened the "USS Lafayette"), which was the largest ship in the world when she capsized and sank in New York Harbor in 1942. The first chapter discusses the business aspects of being a diver entrepeneur. Topics, such as the following, are well covered: legal and insurance requirements, recordkeeping, the Carpenters and joiners union, advertising, contracts, bidding, and how to set a price for one's services. Other chapters are devoted to equipment and tools, search and recovery, ship repair, salvage, pile driving and removal, underwater cutting and welding, the use of explosives, underwater concrete construction, and the laying of pipe and cables. The practical information includes, for example, instructions in how to construct airlifts, patches, ladders, and discusses various types of jetting nozzles in the Equipment, Tools and Methods chapter, as well as the advantages and disadvantages of SCUBA versus various surface-supplied diving equipment for different jobs. Most of the photographs are from U.S. Navy sources and are of good quality (all are black and white). The sketched illustrations are often quite crude and are sometimes somewhat blotchy or smeared. FHowever, they are quite understandable to the experienced diver, and so the artistic deficiency does not detract from the utility of the illustrations. I regard this book to be a valuable asset to not only the commercial diver's library, but also to Naval and research organizations (and, I suspect, it will be irresistable to experienced sport divers despite the caveats of the preface). It is interesting to note that, despite the high-technology developments which have revolutionized the diving industry, old fashioned ingenuity and common sense have not gone out of style. Best Pubiishing's Diver's Reference Dictionary is a handy lexicon of terms used in the diving industry. Divers, like all other occupations, have developed a rich jargon with special meaning. The book also contains definitions of terms from the various sciences and the medical field useful to divers, as well as an extensive table of conversion factors as an appendix. There are 105 conversion factors listed for the unit of pressure ."atmosphere," for example. (I found it curious, however, that, although there were several conversion factors from atmosphere to inches, feet, meters, etc. of water at 60 degrees Fahrenheit, no entries were made for seawater, which is a standard unit in the diving industry). The definitions could have been more complete as well. For example, "diaphragm" was correctly defined in the anatomical sense, but no mention was made of the importance of the term in the mechanical sense, as in regulators and other pressure sensing devices. There also are no illustrations, which would have greatly increased the utility of this book. It would be helpful, for example, to be able to look at an illustration of a "christmas tree," a "dip meter," or a "devil's claw," as well as read the description. Terrence M. Rioux, Diving Safety Officer, Woods Hole Oceanographic Institution The Marine Environment of the U.S. Atlantic Continental Slope and Rise, John B. Milliman and W. Redwood Wright, eds. 1987. Jones and Bartlett Publishers, Inc., Boston, MA. 275 pp. + viii. $50.00. The Atlantic continental slope and rise is perhaps the best studied outer continental margin in the world. Yet with the exception of the monograph prepared in 1972 by K. O. Emery and Elazar Uchupi, no attempt has been made to summarize this extensive and diverse environment. In their book, Milliman and Wright condense a report prepared for the Minerals Management Service of the U.S. Department of the Interior in 1984 by Marine Geoscience Applications, Inc. (Woods Hole, Mass.). In addition, they examine three general areas in which new data must be acquired if we are to improve our understanding of the slope and rise. A detailed bibliography is listed at the end of the book for those interested in further readings. Titanic Declared a Maritime Memorial Legislation to declare the Titanic an international maritime memorial was signed by the President in late October, 1986. The bill (S.2048): declares the site of the shipwreck an international maritime memorial; commends the U.S. -French expedition which discovered the wreck; directs the National Oceanic and Atmospheric Administration to develop guidelines for research, exploration, and, if appropriate, salvage of the wreck; directs the State Department to begin negotiations for an international agreement to manage the wreck; and, expresses the sense of the Congress that the wreck not be disturbed until this agreement is in place. The wreck was located in international waters on September 1, 1985 by a joint U.S.- French team led by Robert Ballard of the Woods F-lole Oceanographic Institution. The ship sank near midnight on April 14, 1912. (See Oceanus Vol. 28, No. 4, and Vol. 29, No. 3.) 95 The History of Whaling in the Western Arctic JOHN R. BO whales. Ice, and Men — The History of Whaling in the Western Arctic, by John R. Bockstoce. 1986. Published by the University of Washington Press, Seattle, Washington. 400 pp. $29.95. "On July 23, 1848, Captain Thomas Roys stood on the deck of the whaling bark Superior with a revolver in hand. He had concealed his course from the crew, but upon discovering their location, they were consumed by a fright that drove his first mate to tears." Their location — the Bering Strait, the shallow, 50- mile-wide gut that separates the Pacific from the Arctic Ocean. Believing that there might be money to be made beyond the Bering Strait, he had ventured $100 to purchase the Russian charts of those waters. "When the fog lifted, Roys knew that he had made the greatest whaling discovery of the century." And, by this time, the reader knows that he/she has discovered a truly fine book. A book whose interest goes far beyond whaling, since, as the author states, and later fully documents, "Roys's cruise was not only the most important whaling discovery of the 19th century, it was also one of the most important events in the history of the Pacific." Bockstoce quickly sets the tone of the book, and sets aside the romantic rubbish one frequently encounters related to 19th century whaling: "Whalemen did not set out on four-year voyages for reasons of national expansion, geographical discovery, religious proselytization, debauchery, scientific inquiry, or a number of other motives attributed to them by modern writers. Simply put, most writers have ignored the industry's essential element — everyone involved entered it solely to make a profit. The whaling industry simply represented the Jndustrial Revolution's expansion to the most remote waters of the world." Once into the northern Bering Sea, the whalers were in the Arctic. The Bering Sea is the world's third largest, but it supports one of the richest ecosystems. It is nourished by the ocean currents which sweep north out of the Pacific, rise over the continental shelf, and carry nitrates, phosphates, and other nutrients. These mix with the oxygen-rich upper waters and produce a fertile broth. As the ice recedes in the lengthening Arctic days, the action of sunlight triggers a brief but intense plankton bloom. For the most part, the bloom occurs at the edge of the melting pack ice. Here, the whalemen hunted the bowheads grazing the lush marine garden in the 24-hour daylight. As the ice receded, the pursuit continued northward beyond the Bering Strait and into the Chukchi and Beaufort Seas. It was far from idyllic. Very far. With warm winds passing over cold water, the whalers' main problem was fog. Close behind were the ship-crushing ice, frostbite cold, treacherous shoals, howling gales, and strong currents. Upon leaving the Arctic, one logbook recorded, "Bound to the south and God forbid that we ever come back again." Early on, as early as five years after the fishery began, there were signs that the western Arctic had been "fished-out." A few raised questions about the possible extermination of the whales. Bockstoce observes, "The whalemen were, however, held captive by the same forces that drive all boomers . . . those who show forbearance are merely those who profit less." Indeed the profit was sometimes hard to come by, and even if a whale was struck, there was no guarantee. Many escaped. As harpoons, or irons, gave way to various guns, the bomb-lance, fired from a shoulder gun, was developed. To the gunner, standing in the bow of the whaleboat, "It gave a brutal kick — enough to throw the man over backwards." It was known to "kill at one end and wound at the other." One captain is quoted, "it was a marvelous creation, almost as deadly at the breech as at the muzzle." It is no small irony that Captain Thomas Roys, the man who led the hunters to the whales, lost his left hand in 1858 in an explosion while experimenting with a rocket harpoon. Throughout the book, the often complex interaction of the whalers with the natives is described. These relationships were often mutually beneficial. Sometimes they were not. Walrusing is a case in point. The natives depended on the walrus herds, not the whales, as their most reliable food resource. Shortly after arriving, the whalers discovered the walrus. From 1869- 1878 more than 100,000 walruses were taken, in what can only be described as a slaughter. Aside from the walruses, it was the natives who suffered most. Hardship, starvation, death, and the abandonment of whole villages paralleled the depletion of the walrus herds by the whalemen. Some natives who survived were reduced to eating their dogs and boat covers. One aspect of Arctic whaling unknown to many, but described in this book, is the trading that took place. For some, the revenues from trading were as important as the whaling, and sometimes trading was indeed the primary mission. One passage in particular elicited a smile: "Chicanery was part of trading in the Bering Strait, although it was practiced by comparatively few. Some captains carefully poured alcohol into containers that were partly full of water so that the alcohol rested on the surface. Others gave out samples of full-strength whiskey and then sold a heavily diluted mixture. In return, it was relatively common to find fox tails sewn on rabbit skins, or damaged fox skins cleverly patched with rabbit fur; broken walrus tusks riveted together with lead and the joints concealed by smeared reindeer fat; and stones set in the root canals of walrus tusks to increase their weight." 96 In the mid-1870s, fashion began the trend to a wasp-waist silhouette. The demand for corset bones and busks drove the price of whalebone up — and drove the whalemen deeper into the Arctic. It is here where perhaps the most gripping part of the book is located. It is the era of steam whaling, the Pacific Steam Whaling Company, and — the overwintering of ships and whalemen in places like Point Barrow, Point Hope, Herschel Island, and Baillie Island. From the end of one season to the beginning of the next, through the long, dark, Arctic winter, whalemen, sometimes their families, and often a retinue of hired natives waited — waited for the bowheads to return. On occasion, it happened that it was not until after the 4fh of July that the ships were freed from the ice and the whaling began. The photographs and illustrations in the book complement the quality of the text. They are equally excellent and vivid — and numerous. The maps on the endsheets are a nice touch. The backmatter, like the book, is complete and scholarly. There are 5 appendices (including a glossary), notes and references listed by chapter, a bibliography, and, in my mind at least — the hallmark of a fine book — a complete and useful index. I have only a single quibble with this fine book. On page 27, Bockstoce defines spermaceti as the whaler's term for a waxy substance found in the sperm whale's head. In fact, spermaceti is not confined to the head matter of the sperm whale, but is found throughout the oil. However, the head oil or matter gave a greater proportion of spermaceti, and was thus more valuable to candlemakers than the body oil. For this reason, the two oils were often stored and marketed separately. Ail in all, Bockstoce has collected together into one book a complete telling of the story of Arctic whaling. I learned a great deal, and in a thoroughly enjoyable way. In short, it is an excellent book, a great read, and a highly recommended addition to the library of those with interests in whaling, history, exploration, and anthropology. James H. W. Hain, Assistant Editor Oceanus Books Received Biology Bulletin of the Biological Society of Washington, Meredith L. )ones, ed. 1985. Published for the Biological Society of Washington by INFAX Corp., Vienna, VA. 547 pp. + v. $30.00 A Monograph on Polyclad Turbellaria by Stephen Prudhoe. 1986. Oxford University Press, New York, NY. 259 pp. $65.00 Marine Biology: Environment, Diversity, and Ecology, by Matthew Lerman. 1986. The Benjamin/ Cummings Publishing Co., Inc., Menio Park, CA. 534 pp. + xiv. $28.95 Marine Interfaces Ecohydrodynamics, Jacques C. J. Nihoul, ed. 1986. Elsevier Science Publishing Co., Inc., New York, NY. 670 pp. + xiv. $79.75. Plankton Dynamics of the Southern California Bight, Richard W. Eppley, ed., 1986. Springer- Verlag, New York, NY 373 pp. + xiii. $31.40. Reptiles — Their Latin Names Explained by A.F. Gotch. 1986. Blandford Press, New York, NY. 176 pp. $24.95. Sea Ice Biota, Rita A. Horner, ed. 1985. CRC Press, Inc., Boca Raton, PL. 215 pp. $90.00 Climate Climatological Atlas of the World Ocean, Sydney Levitus. 1982. U.S. Department of Commerce, NCAA, Washington, D.C. 173 pp. + xiii. $11.00. Climatic Change & World Affairs by Crispin Tickell. 1986. University Press of America, Lanham, MD. 76 pp. + xii. $7.25. Diving Scuba Northeast, Volume II, by Robert G. Bachand. 1986. Sea Sports Publications, Norwalk, CT. 151 pp. + vii. $8.95. Engineering 7985 Australasian Conference on Coastal and Ocean Engineering. 1985. Sponsored by the Institution of Engineers, Australia; Institution of Professional Engineers, New Zealand; and National Water and Soil Conservation Organisation. Volumes 1 and 2. 1174 pp. $66.00 (both volumes) Floating Ports: Design and Construction Practices by Gregory P. Tsinker. 1986. Gulf Publishing Co., Houston, TX. 380 pp. + xvi. $48.00. Environment/Ecology Estuarine Cohesive Sediment Dynamics, Ashish J. Mehta, ed. 1986. Springer- Verlag, New York, NY. 473 pp. $40.00. Pacific Mineral Resources: Physical, Economic, and Legal Issues, Charles J. Johnson and Allen L. Clark, eds. 1986. Sponsored by The East West Resource Systems Institute, Honolulu, HI. 639 pp. -t- xvii. $40.00. Physics of Shallow Estuaries and Bays, J. van de Kreeke, ed. 1986. Springer-Verlag, New York, NY. 280 pp. + vii. $24.50. The Role of Freshwater Outflow in Coastal Marine Ecosystems, Stig Skreslet, ed. 1986. Springer-Verlag, New York, NY. 453 pp. + xi. 97 Sedimentation and Mineral Deposits in the Southwestern Pacific Ocean, D. S. Cronan, ed. 1986. Academic Press, New York, NY. 344 pp. + ix. $67.00. Strategies and Advanced Techniques for Marine Pollution Studies: Mediterranean Sea, C.S. Giam and H.J.-M. Dou. 1986. Springer- Verlag, New York, NY. 475 pp. + xii. $89.00. Tidal Mixing and Plankton Dynamics, Malcolm ). Bowman, Clarice M. Yentsch, and William T. Peterson, eds. 1986. Springer- Verlag, New York, NY. 502 pp. + x. $59.00. Turbulence and Diffusion in Stable Environments, J.C.R. Hunt, ed. 1985. Oxford University Press, New York, NY. 319 pp. + xv. $49.00. Variability and Management of Large Marine Ecosystems, Kenneth Sherman and Lewis M. Alexander, eds. 1986. Westview Press, Inc., Boulder, CO. 319 pp. + xxvi. $31.95. Fisheries The Economics of Fisheries Management by Lee G. Anderson, 1986. The Johns Hopkins University Press, Baltimore, MD. 296 pp. + xx. $29.95. General Reading Cannibals and Condos: Texans and Texas along the Gulf Coast by Robert Lee Maril. 1986. Texas A&M University Press, College Station, TX. 113 pp. + XV. $13.95. Dean's Doctrine of Naval Architecture 1670, Brian Lavery, ed. 1986. Naval Institute Press, Annapolis, MD. 128 pp. $22.95. Exploring the Oceans: An Introduction for the Traveler and Amateur Naturalist by Henry S. Parker. 1985. Phalarope Books, Prentice-Hall, Englewood Cliffs, N|. 354 pp. + xiv. $15.95. The Future of the Oceans by Elisabeth Mann Borgese. 1986. Harvest House, Ltd., Montreal, Canada. 144 pp. + xvi. $9.95. Icebound: The Jeannette Expedition's Quest for the North Pole by Leonard F. Guttridge, Naval Institute Press, Annapolis, MD. 357 pp. + XX. $23.95. The Last Extinction, Les Kaufman and Kenneth Mallory, 1986. The MIT Press, Cambridge, MA. 208 pp. + ix. $16.95. Nicholas Pocock 1740-1821 by David Cordingly. 1986. Naval Institute Press, Annapolis, MD. 120 pp. $18.95. PBY/The Catalina Flying Boat by Roscoe Creed. 1985. Naval Institute Press, Annapolis, MD. 351 pp. + xvii. $21.95. Sevengill: The Shark and Me by Don. C. Reed. 1986. Sierra Club Books, San Francisco, CA. 125 pp. + ix. $11.95. Geology Submarine Fans and Related Turbidite Systems, A.H. Bouma, W.R. Normark, and N.E. Barnes, eds. 1985. Springer-Verlag, New York, NY. 351 pp. + xiv. $59.00. History Between Pacific Tides by Edward F. Ricketts, Jack Calvin, and Joel W. Hedgpeth. 1985. Stanford University Press. Stanford, CA. 652 pp. + xxvi. $29.50. Marine Policy Antarctic Treaty System: An Assessment. 1985. National Academy Press, Washington, DC. 435 pp. + XV. $22.50. International Law for Seagoing Officers by Burdick H. Brittin. 1986. Naval Institute Press, Annapolis, MD. 503 pp. + XV. $24.95. Offshore Lands: Oil and Gas Leasing and Conservation on the Outer Continental Shelf by Walter |. Mead, Asbjorn Moseidjord, Dennis D. Muraoka, and Philip E. Sorensen. 1985. Pacific Institute For Public Policy Research, San Francisco, CA. 169 pp. + xxviii. $12.95 (paperback), $34.95 (hardcover). Proceedings of the 1985 California Offshore Petroleum Conference, W.N. Tiffney, Jr. 1985. Pallister Resource Management, LTD. Malibu, CA. 259 pp. + xxi. $75.00. Ships and Sailing The Art of Knotting and Splicing by Cyrus Lawrence Day. 1986. Naval institute Press, Annapolis, MD. 235 pp. + xiv. $19.95. Combat Fleets of the World: 1986/ 87 Their Ships, Aircraft, and Armament, Jean Labayle Couhat, ed. 1986. Naval Institute Press, Annapolis, MD. 750 pp. + xii. $94.95. Command Under Sail: Makers of the American Naval Tradition, 1775-1850, James C. Bradford, ed. 1985. Naval Institute Press, Annapolis, MD. 333 pp. -I- xvi. $24.95. Formulae for the Mariner by Richard M. Plant. 1986. Cornell Maritime Press, Centreville, MD. 93 pp. + xiv. $12.50. Handbook of the Nautical Rules of the Road by Christopher B. Liana and George P. Wisneskey. 1986. Naval institute Press, Annapolis, MD. 223 pp. + xii. $16.95. HMS Beagle, 1820-1870: Voyages Summarized, Research and Reconstruction by Lois Darling. 1984. National Maritime Historical Society. Croton-on-Hudson, NY. 12 pp. $1.75. Merchantman? Or Ship of War by Charles Dana Gibson. 1986. Ensign Press, Camden, ME. 214 pp. + ix. $18.75. North to Thule by John and Harriet Frye. 1985. Algonquin Books of Chapel Hill, Chapel Hill, NC. 190 pp. + xxiv. $16.95. Surveying and Charting of the Seas by W. Langeraar. Elsevier Oceanography Series, 37. 1984. Elsevier, New York, NY. 612 pp. + xvii. $53.75. Ship and Boat Models in Ancient Greece by Paul Forsythe Johnston. 1985. Naval institute Press, Annapolis, MD. 187 pp. + xii. $24.95. Sound Sonar Images by Harold E. Edgerton. 1986. Prentice-Hall, Englewood Cliffs, N|. 296 pp. $24.95 (paperback), $36.95 (hardcover). 98 Name: Address: Send $5.00 and your mailing address to: Oceanus Subscription Service Center P.O. Box 6419 Syracuse, NY 13217 City, State, Zip: Make checks payable to: Woods Hole Oceanographic Institution The Titanic Reuisited the story of the second expedition to the grave site in the summer of 1986 also available for $4 INDEX VOLUME 29 (1986) Number 1, Spring, The Arctic Ocean: James H. Zumberge, Introduction — Oran R. Young, The Age of the Arctic — Norbert Untersteiner, Glaciology — A Primer on Ice — Thomas Newbury, Sea Ice and Oceanographic Conditions — Vera Alexander, Arctic Ocean Pollution — Maxwell ]. Dunbar, Arctic Marine Ecosystems — D. James Baker, The Arctic's Role in Climate — Lawson W. Brigham, Arctic Icebreakers — U.S., Canadian, Soviet — Wiltord F. Weeks and Frank D. Carsey, Remote Sensing of the Arctic Seas — Dean A. Horn and G. Leonard Johnson, MIZEX East: Past Operations, Future Plans — James W. Curlin, Peter Johnson, William Westermeyer, and Candice Stevens, Arctic Offshore Petroleum Technologies — Paul R. Ryan, Concerns: Soviets Shelve Plan on Diverting Rivers in Arctic Regions — Tina Berger, Concerns: Bowhead Estimates Revised Upward; Hunt Issues Ease — Frank Lowenstein, Profile: Judith M. Capuzzo: Educating the Guessers — Sarah A. Little, Cruise Report: First Argo Science Test Yields Unexpected Data on Ridge — Letters — Book Reviews Number 2, Summer, The Great Barrier Reef: Science and Management: Barry O. Jones, Introduction: The Great Barrier Reef Science & Management — David Hopley, and Peter J. Davies, The Evolution of the Great Barrier Reef — Graeme Kelleher, Managing the Great Barrier Reef — David J. Barnes, Bruce E. Chalker, and Donald W. Kinsey, Reef Metabolism — J. E. N. Veron, Distribution of Reef-Building Corals — John C. Coll and Paul W. Sammarco, Soft Corals: Chemistry and Ecology — Garden C. Wallace, Russell C. Babcock, Peter L. Harrison, James K. Oliver, and Bette L. Willis, Sex on the Reef: Mass Spawning of Corals — Llewellya Hillis- Colinvaux, hiistorical Perspectives on Algae and Reefs: tiave Reefs Been Misnamed^ — Michael A. Borowitzka, and Anthony W. D. Larkum, Reef Algae — John Lucas, The Crown of Thorns Starfish — Photo Essay: Images From the Underwater Outback — Clive R. Wilkinson, The Nutritional Spectrum in Coral Reef Benthos — or Sponging Off One Another for Dinner — David McB. Williams, Garry Russ, and Peter J. Doherty, Reef Fish: Large-Scale Distribution and Recruitment — Eric Wolanski, David L. B. Jupp, and George L. Pickard, Currents and Coral Reefs — D. A. Kuchler, Remote Sensing: What Can It Offer Coral Reef Studies? — Harold Heatwole, and Peter Saenger, Islands and Birds — Brydget E. T. Hudson, Dugongs and People — M. K. James, and K. P. Stark, Risk Analysis: Cyclones, and Shipping Accidents — J. T. Baker, and J. A. Williamson, Toxins and Beneficial Products from Reef Organisms — Research Stations on the Great Barrier Reef: Lizard Island, One Tree Island, Orpheus Island, Heron Island — Barbara E. Kinsey, Profile: Joseph T. Baker, Early Man (3 a.m.) — Book Reviews Number 3, Fall, The Titanic Revisited, the Irish Sea, Aquariums, and Ocean Architecture: Paul R. Ryan, The Titanic Revisited — James H. W. Hain, Low-Level Radioactivity in the Irish Sea — Eleanore D. Scavotto, Research Plays Key Role in Growth of U.S. Aquariums — Philip Rabinowitz, Sylvia Herrig, and Karen Riede!, Ocean Drilling Program Altering Our Perception of Earth — Charles N. Ehler, Daniel J. Basta, Thomas F. LaPointe, and Maureen A. Warren, New Oceanic and Coastal Atlases Focus on Potential FEZ Conflicts — Paul R. Ryan, and Michael A. Champ, Ocean Architecture and Engineering in japan — Paul Ferris Smith, Dodge Morgan, the Argos System, and Oceanography — James H. W. Hain, URI Symposium Report: The Future of the World's Oceans — Lauriston R. King, Concerns: Staying Alive — Sea Grant and the Budget Battle — Sally Ann Lentz and Clifton E. Curtis, Concerns: EPA Puts the Ice on Ocean Burns — Timothy Eichenberg, Concerns: California Oil Case Tests State-Federal Coastal Role — Bruce Finson and Katherine E. Taylor, History: Steinbeck & Ricketts: Fishing in the Mind — Book Reviews Number 4, Winter, Changing Climate and the Oceans: Francis P. Bretherton, Introduction: The Oceans, Climate, and Technology — Berrien Moore III and Bert Bolin, The Oceans, Carbon Dioxide, and Global Climate Change — James J. McCarthy, Peter G. Brewer, and Gene Feldman, Global Ocean Flux — Meinrat O. Andreae, The Oceans as a Source of Biogenic Gases — Kirk Bryan, Man's Great Geophysical Experiment: Can We Model The Consequences? — Nicklas G. Pisias and John Imbrie, Orbital Geometry, Carbon Dioxide, and Pleistocene Climate — Stanley S. Jacobs, The Polar Ice Sheets: A Wild Card in the Deck? — Julie M. Palais, Polar Ice Cores — Photo Essay: Humans in Ice Age Europe — 10,000 to 35,000 Years Ago — Linda E. Heusser, Pollen in Marine Cores: Evidence of Past Climates — George M. Woodwell, Forests and Climate: Surprises in Store — D. James Baker and W. Stanley Wilson, Spaceborne Observations in Support of Earth Science — James H. W. Hain, Profile: Robert A. Frosch: Unafraid to Take Risks — Book Reviews — Index 100 Give Gift of the Sea cv^«*°p«^>,. This Season come Oceanus The International Magazine of Marine Science and Policy Published by Woods Hole Oceanographic Institution Foreign Subscription Order Form: Outside U.S. & Canada' Please make cheques payable to Cambridge University Press D one year at £20.00 Library or Institution: D one year at £37.00 D payment enclosed, (we require prepayment) Please send MY Subscription to: Please send a GIFT Subscription to: Name (please print) Name (please print) Street address Street address City State Zip City State Zip *U.S. and Canadian subscribers please use form inserted at Donor's Name, front of issue. Address 12/86 Vol 27:3, Fall 1984- A lull report on vent science. • El Nino, Vol 27:2, Summer 1984 —An atmospheric phenomenon analyzed. • Industry and the Oceans, Vol. 27:1, Spring 1984 • Oceanography in China, Vol. 26:4, Winter 1983/84 • Offshore Oil and Gas, Vol. 26:3, Fall 1983 p^ou, vol Zi.z — nanKion, ci tNmo ana Airitan tisneries, noi springs, i^eorges Bank, and more • A Decade of Big Ocean Science, Vol. 2 3:1, Spring 1980. • Ocean Energy, Vol 22:4. Winter 1979/80. • Ocean/Continent Boundaries, Vol. 22:3, Fall 1979 • Sound in the Sea, Vol 20:2, Spring 1977 - The use of acoustics in navigation and oceanography. Issues not listed here, including those publfshed prior to 1977, are out of print They are available on microfilm through University Microfilm International, 300 North Zeeb Road, Ann Arbor, Ml 48106 Back issues cost $4.00 each, except for Great Barrier Reef and Titanic issues, which are $5 There is a discount of 25 percent on orders of five or more. Orders must be prepaid; please make checks payable to Woods Hole Ocean- ographic Institution. Foreign orders must be accompanied by a check payable to Oceanus for £3.50 per issue (or equivalent) Send orders to: Oceanus back issues Subscriber Service Center P.O. Box 6419 Syracuse, NY 13217 HAS THE SUBSCRIPTION COUPON BEEN DETACHED? If someone else has made use of the coupon attached to this card, you can still subscribe. Just send a cheque — £20 for one year (four issues) — to this address: Please make cheques payable to Cambridge University Press ,^^»*Of*.,. 1930 Cambridge University Press The Edinburgh Building Shaftesbury Road Cambridge CB2 2RU England I ss ■ s ■ a I PLACE STAMP HERE Oceanus Cambridge University Press The Edinburgh Building Shaftesbury Road Cambridge CB2 2RU England L-oncerns: (^aiiiornia kju cdie /eiti otdte-reuerd; v„L/a>id( /\u/tr History: Steinbeck & Ricketts: Fishing in the Mind— Book Reviews L)l UV_C I III5WII aiiu ixaiiidiinr l. layiwi, Number 4, Winter, Changing Climate and the Oceans: Francis P. Bretherton, Introduction: The Oceans, Climate, and Technology— Bernen Moore III and Bert Bolin, The Oceans, Carbon Dioxide, and Global Climate Change— \ames J. McCarthy, Peter G. Brewer, and Gene Feldman, Global Ocean f/ux— Meinrat O. Andreae, The Oceans as a Source of Biogenic Cases— Kirk Bryan, Man's Great Geophysical Experiment: Can We Model The Consequences^— N'\ck\as G. Pisias and John Imbrie, Orbital Geometry, Carbon Dioxide, and Pleistocene Climate-Stanley S. Jacobs, The Polar Ice Sheets: A Wild Card in the Dec/