Global Environmental Change: Research Pathways for the Next Decade, Overview (1998)

OVERVIEW

Global Environmental Change: Research Pathways for the Next Decade

During its first 10 years of operation, the United States Global Change Research Program (USGCRP) has advanced our understanding of the Earth's ever-changing physical, chemical, and biological systems, and the growing human influences on these systems. On the basis of this knowledge, we can now focus attention on the critical unanswered scientific questions that must be resolved to fully understand and usefully predict global change. Such capability is increasingly important for developing our economy, protecting our environment, safeguarding our health, and negotiating international agreements to ensure the sustainable development of our nation and the global community of nations. There are now compelling reasons for scientific knowledge to guide and respond to policy options, both current and future. Clearly, we must delineate research pathways that will enlarge our understand ing of changes in the global environment, including climate change. At the same time, we need to reduce uncertainties in the projections that shape our decisions for the future. For all these reasons, it is essential that the USGCRP continue to receive strong financial support and continue to provide continuing strong scientific leadership. To be effective, the USGCRP must be based on a sound scientific strategy, focused on key unanswered scientific questions, using a correspondingly balanced strategy for supporting observational, data management, and analysis activities.

On the basis of the continuing reviews of the Committee on Global Change Research (CGCR) and those of its collaborating bodies, the Committee reaffirms the achievements and significance of the USGCRP while finding that the Program must now be revitalized, focusing its use of funds more effectively on the principal unanswered scientific questions about global environmental change. This goal demands that funding and efforts be directed toward a coherent and coordinated suite of research activities and supporting observational, data management, and modeling capabilities, all aimed at imperative re search objectives and clearly defined scientific questions. A sharply focused scientific strategy and a coherent programmatic structure are both critically needed. This report seeks to provide a framework for such a strategy and structure. The elaboration and implementation of this scientific strategy and programmatic structure will be the principal challenge for global change research over the course of the next decade.

Long before the industrial revolution, human activity began to alter the Earth's environment. However, only in this century has the scale of such alterations become global in scope; moreover, the rate of these recent changes is enormously high compared with the historical record. Today, on the threshold of a new millennium, it is clear that humans are inducing environmental changes in the planet as a whole. In fact, the human fingerprint is abundantly seen on the global atmosphere, the world oceans, and the land of all continents. This insight has brought about profound changes in the goals, priorities, and processes of both science and government.

The recognition that humans are causing global changes in the biology, physics, and chemistry of the environment—changes with immense significance for human society and economy—prompted the US government, and other national governments, to act. In 1990, Congress established the USGCRP to carry out an organized, coherent attack on the scientific issues posed by global environmental change.

The USGCRP had its principal roots in the 1980s, as both scientists and the public became increasingly aware of the links among human activities, current and future states of the global environment, and human welfare. The most immediate concerns were human-induced climate change, stratospheric ozone depletion due to industrial emissions, and emerging evidence that the Earth's biogeochemical system was being perturbed by a broad range of human actions.

Some of the many antecedents of the USGCRP were seen still earlier. In the 1970s, a convergence of long-standing scientific concerns (see below) and a series of climatic events led to the first World Climate Conference and to the establishment of the US National Climate Program and the World Climate Program.¹

In parallel, beginning in the mid-1970s, the US Department of Energy (DOE) organized a major research program to assess the consequences of fossil-based energy production. Workshops chaired by the late Roger Revelle outlined a broad multidisciplinary research agenda closely congruent with today's USGCRP, including a strong emphasis on the carbon cycle, the role of ecosystems, and human dimensions research.²

The immediate precursor of the USGCRP, however, was a workshop sponsored by the National Aeronautics and Space Administration (NASA) in 1982 on global habitability, which was led by Richard Goody.³ This workshop emphasized the fact that, in many critical respects, the ocean, atmosphere, and biosphere function together on long timescales as a single integrated system, a system requiring interdisciplinary research and observing programs of global scope and decadal duration. The stage had been set for encouraging similar fully integrated, long-term research by the Global Atmospheric Research Program, a program that itself arose from a seminal study by the National Research Council (NRC)⁴ and laid the groundwork for the World Climate Research Program. The shaping of such comprehensive endeavors, which arose by recognizing the importance of chemical and biological as well as physical factors in the global system, also led to the establishment of the International Geosphere-Biosphere Program of the International Council of Scientific Unions. The priorities and nature of this program, from a US perspective, were laid out in a sequence of NRC reports.⁵ Most recently, human components in global environmental change have been given wider recognition in the creation of the International Human Dimensions Program on Global Environmental Change.

Still other precursors to the USGCRP include two reports in the 1980s by the NASA-sponsored Earth System Sciences Committee (ESSC),⁶ which sought to define a new and revolutionary scientific discipline of Earth System Science. In keeping with the Goody Report⁷ and the 1986 NRC report, Global Change in the Geosphere-Biosphere,⁸ this new discipline would be dedicated to study of the Earth as an integrated system of interacting components. Its goal would be to obtain “a scientific understanding of the entire Earth System on a global scale.”⁹ The emergence of a science of the Earth system, moreover, offered a promise of knowledge that would be valuable to decision makers addressing global habitability.

Prominent in the ESSC documents was a recommendation for an Earth Observing System (EOS) to provide long-term global observations, with an empha-

sis on the long-term continuity of observations, both satellite and in situ. The importance of long-term records reflected the audience for these reports and portended a multiagency endeavor: the recommendations were made to several concerned agencies—to the National Oceanic and Atmospheric Association (NOAA) and the National Science Foundation (NSF)—in addition to the sponsoring agency, NASA.

Late 1986 brought the beginnings of a coordinated government response. NASA, NOAA, and NSF had been developing parallel global change programs, but in 1987, a joint letter from these three agencies to the director of the Office of Management and Budget (OMB) proposed the idea of a budget presentation coordinated across the agencies. From this point on, OMB was instrumental in developing the USGCRP. Later that year, a consortium of eight agencies formed the federal interagency Committee on Earth Sciences (later the Committee on Earth and Environmental Sciences, now the Committee on Environment and Natural Resources). The first funding for the USGCRP per se came in fiscal year 1989, and the first related descriptive document that accompanied the President 's Budget was produced for the fiscal year 1990 submission. Joint submission of agency budgets was a novel concept, at least in the Earth sciences. The process produced new initiatives that were coordinated, if not necessarily integrated. Thus, the USGCRP was initiated and first presented in the federal budget by President Reagan, was codified into law in 1990 (see Appendix A), and was implemented by President Bush; today it is being carried forward under President Clinton.

The intellectual crucible in which the USGCRP was formed, however, was itself forged far earlier. The possibility of global changes in the biological, physical, and chemical environment had been recognized in the 19th century and became a widely accepted idea by the beginning of the 20th century. In 1957, Revelle and Suess¹⁰ pointed out that most of the carbon dioxide emitted from fossil fuel combustion would remain in the atmosphere for many years and drew on emerging climate modeling capabilities to suggest possibly alarming impacts on climate. In the early 1960s, two major international conferences, known by the acronyms SMIC and SCEP,¹¹ put the issue on the international agenda. At the same time, convincing observational evidence emerged that human activities were in fact changing the chemical composition of the global atmosphere. Measurements taken first by Charles David Keeling in 1957 revealed that carbon dioxide was indeed increasing in the atmosphere at the planetary scale. In 1964, the President's Science Advisory Council brought the issue to the attention of the US Government. Subsequently, beginning in the late 1960s, early computer model simulations started to explore the possible changes in temperature and precipitation that could occur due to increasing human-induced emissions of greenhouse gases into the atmosphere.

During the 1970s and early 1980s, an important set of environmental topics was closely considered by the National Academy of Sciences (NAS). Foremost among these issues were potential changes in climate and losses in stratospheric ozone. The NAS convened several panels and committees under leading scientists such as the late Roger Revelle ¹² and Jule Charney.¹³ The resulting reports projected that energy production from fossil fuels would continue to increase atmospheric concentrations of carbon dioxide and estimated that a doubling of the atmosphere's carbon dioxide concentration could potentially raise global average temperature by 1.5 to 4.5° C (about 2.7 to 8° F) and produce a complex pattern of worldwide climate changes. Charney and his colleagues concluded that if carbon dioxide continued to increase, there was “no reason to doubt that climate changes will result and no reason to believe that these changes will be negligible.”¹⁴ The Revelle group saw a clear need for two kinds of action in response: “organization of a comprehensive worldwide research program and new institutional arrangements.” In the same period, ecologists also recognized that massive changes in ecosystems due to land-use changes and other stresses could affect the carbon cycle. In this juncture of scientific findings, then, are the beginnings of the partnerships among the life and Earth sciences that have become the hallmark of global change science.

Still other studies addressed a widening range of potential global change impacts and their policy implications.¹⁵ In 1979 and 1989, major World Climate Conferences¹⁶ were convened by the World Meteorological Organization and other international bodies. International meetings¹⁷ converged on the conclusion that the implications of changing climate should be assessed for development policy. In 1988, the Intergovernmental Panel on Climate Change, composed of hundreds of scientists from more than 50 countries, assumed responsibility for conducting periodic international assessments on climate change and its consequences. The latest of these¹⁸ affirms the validity of scientific concerns and concludes that human influences on climate are becoming discernible.

Thus, throughout the last two decades, the NAS/NRC and their international counterparts have continued to examine the science of climate change and variability and the associated policy implications for the United States and other nations. Additionally, the NAS/NRC has simultaneously considered climate change and variability within the broader context of global change. The Committee on Global Change Research (CGCR), author of this report, and CGCR's predecessor, the Board on Global Change, have been charged with providing continuing guidance to national and international global change efforts. In 1995, CGCR undertook an initial assessment of the scientific programs of the USGCRP, reviewed the specific role of NASA's Mission to Planet Earth/Earth Observing System (MTPE/EOS), and issued a report with recommendations (the “La Jolla” report)¹⁹ and a follow-up report on the government response.²⁰ The present study significantly expands that effort.

A related history of research concerns another pressing environmental issue—the depletion of the stratospheric ozone layer that shields us from damaging ultraviolet radiation. In the early 1970s, proposals to build a fleet of supersonic transports raised questions about possible damage to the ozone layer from engine emissions in the stratosphere. A major US research and assessment program was launched, and the NRC was commissioned to conduct a series of studies.²¹ But soon, Rowland and Molina made the startling discovery that chlorofluorocarbons (CFCs), not airplanes, were the frightening threat to our ozone shield. Eventually, an international assessment was conducted under the auspices of the World Meteorological Organization and other international bodies.²²

The discovery of Rowland and Molina^a reminds us that studies and reports often do not adequately address the complexities of the real world. Indeed, they can even significantly miss the mark. Studies of ozone depletion had focused on slow incremental changes and had sought incremental improvements through corresponding models and parametric analyses. Meanwhile, observations extending back to the 1950s had been tracking the amount of ozone over the Antarctic each year through its seasonal cycle. In the late 1970s, an anomalous deficit was observed in the total amount of ozone over the Southern Hemisphere in late winter observations. Then, in 1985, the British Antarctic Survey reported dramatic—and rapidly worsening —ozone losses in springtime ozone concentrations over Halley Bay.

Theories about the cause of this unprecedented and unexpected loss blossomed. Explanations ranged from the hypothesis of the simple redistribution of stratospheric ozone by atmospheric motion to proposed chemical reactions initiated by the magnetic field-focusing of solar electrons and protons. More complete information was clearly needed. In 1986, NASA began planning an airborne expedition using the ER-2 aircraft to penetrate the region of the stratosphere where ozone was disappearing. The mission, executed in August and September 1987 from Punta Arenas, Chile, demonstrated that ozone was being destroyed by chlorine and bromine radicals. The role of CFCs—molecules that transport chlorine to the stratosphere—in the destruction of Antarctic ozone was unequivocally confirmed. Shortly thereafter, laboratory and theoretical work pinned down other essential mechanisms of the process—mechanisms involving cloud particles, which had been overlooked in earlier studies.

With such overwhelming evidence in hand, the nations of the world moved

with remarkable alacrity to mitigate the threat. International meetings developed strategies to control emissions of ozone-destroying substances, while the chemical industry worked to devise substitutes for CFCs. Within a few short years, a comprehensive framework for controlling worldwide emissions had been put in place in the form of the justly admired “Montreal Protocol.”²³

A number of lessons relevant to the broader field of global change research may be drawn from the case of research on Antarctic ozone depletion. The severity of the ozone phenomenon demonstrates that environmental changes are not always incremental or slight. Moreover, the severity of ozone loss came as a total surprise, even though the topic had been carefully considered by the scientific community. Finally, however, the problem was assessed in remarkably short order and effective remedial measures were rapidly instituted—because a solid base of related scientific understanding had been developed through decades of focused observation and research.

An additional critical point to make in this context is that many issues in global environmental change, such as climate change, are far more complex than even the difficult ozone story. The chemical, physical, and biological aspects of the greenhouse problem are extraordinarily daunting to study, and yet an additional, more difficult challenge probably lies in understanding the human dimensions of global change phenomena.

What surprises are in store in the future? By definition, surprises cannot be fully anticipated; at best, they can be acknowledged as possibilities. As such, they pose a special challenge to science. Science must formulate specific questions to set about obtaining the critical observations and performing the analyses needed to answer them. It is hard to ask questions that will anticipate all possible surprises before a surprise occurs.

Preparing science for surprise is, in part, the challenge that the CGCR faced in developing this report. Scientists believe strongly that unfocused research on the complex and varied Earth system is unlikely to be productive. On the other hand, scientists who view the world through pinholes are likely to bump into trees and fall off cliffs. How can needed focus be given to the USGCRP while still casting the research net sufficiently wide to catch the unexpected? In this report, the CGCR has sought to define a framework for this endeavor, identifying a set of coherent domains of research that are likely to provide efficient and productive progress for science and to encompass the range of scientific and social issues implicit in global environmental change. This framework builds on the initial set of guiding principles defined by the Committee in its “La Jolla” report and on the issues of great scientific and practical importance in mature areas of Earth system science that are identified in this report.

This report outlines a research framework across the wide scope of global environmental change in terms of the following primary topical areas:

• Changes in the Biology and Biogeochemistry of Ecosystems

• Changes in the Climate System on Seasonal to Interannual Timescales

• Changes in the Climate System on Decadal to Century Timescales

• Changes in the Chemistry of the Atmosphere

• Paleoclimate

• Human Dimensions of Global Environmental Change

Pathways (see Appendix B) begins with biology and biogeochemistry because of our intimate dependence on biological systems, because of the sensitivity of these systems to changes in the physical and chemical environment, and because of the pivotal role of biology in the changing biogeochemical cycles of the planet. These biogeochemical cycles are, in a sense, the metabolic chart for the planet; they provide particularly useful benchmarks of global change.

We look next into the climate system, focusing initially on climate variability on seasonal to interannual timescales and then on climate change on decadal to century timescales. We find we also must consider climate variability and change on the intermediate timescale of a human generation.

Changes in the chemistry of the atmosphere drive many global changes; the atmosphere quickly transports chemical inputs from whatever source, and the chemical loadings are of sufficient scale that they can no longer be ignored.

Testing ideas about global change on longer timescales is not like research to improve weather forecasts, in which feedback and correction are almost immediate. The paleoclimate record offers a unique opportunity to assess ideas about the dynamics and causes of global environmental change and variability. This record also tells us that large departures from simple expectations have occurred in the past, forcing the recognition that any program addressing global change must be sufficiently broad in scope to ensure that surprises are caught early. This consideration is particularly important for devising observational strategies.

The human dimensions of global environmental change—that is, humans and their institutions as both agents and recipients of change—are integrated where possible into the other topical chapters of this report and are also the subject of a separate treatment. Many concerns about the changing environment are tied directly to concerns about human and ecosystem health and welfare.

The discussion of each of the six primary topical areas is structured in terms of Research Imperatives—central issues posed to the corresponding scientific community by the challenge of global environmental change (see Appendix C). Four to six Research Imperatives are identified for each topical area. Sometimes these imperatives closely interconnect. The Research Imperatives provide the guideposts for the research “pathway.”

Each Research Imperative is addressed by a set of Scientific Questions. The limbs of the research strategy begin to branch and spread. If surprises are in the wind, we hope that this broadly spreading canopy of topics, Research Imperatives, and Scientific Questions will catch the signal.

The Scientific Questions are posed at a level of detail from which an observational program, space-based and in situ, can be defined, refined, and realized. The observational strategy also consciously recognizes that surprises might well be in store. For this and other scientific reasons, an essential requirement of the observational strategy is to establish long-term, scientifically valid, consistent records for global change studies. It is fortunate that the paleoclimate community has provided extremely detailed histories of climate and environmental change that can underpin the instrumental records, establishing some basis for the assessment of future monitoring. Long-term monitoring is a central, scientific challenge for global change research. It is also a difficult challenge to meet in a social environment that so often values or wants something new.

Observations are essential to test hypotheses from which models can be developed. Models are essential if prediction and synthesis are sought. Observations are useless, however, if the data are inaccessible to users (e.g., because of the problem of data recorded in “write-only ” memory). Data systems have been a constant challenge to all scientific investigations; they are particularly problematic when large amounts of data are involved, as in global change studies. Fortunately, through a unique confluence of satellite and computer technology, science stands on the threshold of a greatly enhanced ability to exploit such masses of data and, hence, is well positioned to monitor and predict changes in the global climate and environment. Satellites orbiting the Earth can monitor changes in sea height, wind velocity, atmospheric water vapor, snow cover, and a wide variety of other parameters. Satellite data can be merged with ground-based measurements networks in a matter of minutes through a series of telecommunication satellites, microwave links, and fiber. Data derived from these sources serve as inputs to large computer-based models, which in turn provide predictions about future environmental trends and variability. The existing and future Internet and associated services give the USGCRP an opportunity to manage this stream of data successfully and at reasonable cost.

A data strategy is needed that emphasizes flexible and innovative systems— systems that are less costly than the current EOS core system, that appropriately reflect focused responsibility for data character, that provide open access to the scientific community and the public, and that rapidly track technological developments.

As mandated in the legislation establishing the USGCRP (see Appendix A), the NRC has provided continuing oversight and review of the Program (see Bibliography.

Oversight has been the responsibility of a consortium of NRC groups, coordinated through the former Board on Global Change (now the Board on Sustainable Development and its Committee on Global Change Research, CGCR) and other predecessors. For example, the Climate Research Committee and its panels (operating under the NRC's Board on Atmospheric Sciences and Climate) have overseen climate-related elements of the USGCRP, with particular attention to international programs such as the Tropical Ocean—Global Atmosphere (TOGA) program. The NRC's Committee on the Human Dimensions of Global Change has carried out seminal studies to define social science aspects of the USGCRP. The CGCR and other NRC units receive regular updates on Program status at their meetings. With participation by these and other NRC boards, committees, and panels, the CGCR carried out a comprehensive review of the Program in the summer of 1995,²⁴ which was followed in 1996 by a review of government actions taken in response to the 1995 report.²⁵ In November 1996, the approach to the Pathways report was determined at a CGCR meeting, which for the first time convened representatives from each of the USGCRP agencies and chairpersons and staff of each NRC committee involved in global change research. The findings and recommendations of the present report are based on this continuing stream of review and assessment.

The central purposes of the USGCRP areas follow:

• To observe and document changes in the Earth system

• To understand why these changes are occurring

• To improve predictions of future global changes

• To analyze the environmental, socioeconomic, and health consequences of global change

• To support state-of-the-science assessments of global environmental change issues.²⁶

These “central purposes” of the USGCRP set a clear, appropriate, overarching vision for the Program. Moreover, during the past decade, the USGCRP has realized an impressive array of scientific accomplishments. Progress has been made in understanding the loss of stratospheric ozone, and amendments and adjustments to the Montreal Protocol have benefited from research flowing from the USGCRP. Ice cores have provided evidence of past changes in the Earth's environment, and human-induced environmental changes have been documented. There is a much better understanding, including the development of large-scale models, of the important roles of terrestrial and marine ecosystems in the overall carbon cycle, including knowledge of how such systems might shift under a changing climate. The success in providing predictive, useful information about El Niño–Southern Oscillation (ENSO) phenomena is a significant step in providing scientific information for natural resource management and for improving human welfare, and it offers encouragement that the broader issues of climate variability and human-induced climate change can also be successfully attacked.

Finally, some accomplishments in observations are noteworthy. The precise measurements from space of sea surface height by the US-French Topex-Poseidon mission have advanced our knowledge of sea surface change and ocean circulation. The Mission to Planet Earth Pathfinder data sets have advanced our insights across a wide array of global change issues.

The inherent challenges in achieving the central purposes of the USGCRP, however, will be ongoing; to ensure our well-being for the foreseeable future, it is essential to meet these challenges. They also set a formidable and difficult agenda for science, and this conclusion carries with it the need to do better. We must find ways of advancing the scientific attack on the problems of global environmental change more effectively. Fortunately, with 10 years of experience of successes and setbacks, we are in a far better position to meet the scientific challenges in the coming decade. There is, in fact, a rich body of information, in the form of lessons learned, to be gleaned from the past decade.

What are the lessons of the last 10 years? The reviews carried out over the Program's first decade have in fact identified a key set of “lessons learned”— attributes that the program must maintain and precepts that it must observe to achieve greater and needed successes in attacking the difficult issues of global environmental change.

Need for Programmatic Focus

Where research communities have been given resources based on collaboratively established priorities to implement critical activities, maintain and distribute data sets, and synthesize the information, rapid and impressive progress has been made. Such successes have occurred primarily within the framework of formal programs (e.g., the TOGA studies of El Niño, and the Upper Atmosphere Research Program studies of ozone destruction that led to the Montreal Protocol) and sometimes through grassroots initiatives (e.g., carbon cycle modeling). Many global change projects are currently on a positive trajectory, and success is likely. However, many critical global change questions are not receiving the level of support needed to make similar progress; the sum of support for the current “focused”^b programs, according to the USGCRP specifications, represents an inadequate fraction of what is needed to accomplish its goals. For example, of the

total fiscal year 1998 budget request for the USGCRP, 61% supports space-based observation programs, and 39% supports scientific research. 27

In part, this problem has arisen because of disaggregation of the national effort across multiple agencies. The agencies have neither an enforceable mandate to cooperate in a manner necessary to be successful nor a system that requires accountability of expenditures. The Committee on Environment and Natural Resources (CENR) of the National Science and Technology Council (NSTC) was designed to improve the coordination of both the USGCRP agencies and the budget crosscuts with OMB in presenting a national program. Unfortunately, the management framework has not had the expected effect. The desired “virtual agency”^c has been quite far from reality.

The fact that a principal component^d of the nation's global ocean–carbon cycle research program fell victim to budget reductions during 1996–97 in the DOE and required a last ditch ad hoc rescue by NOAA is a clear statement of programmatic failure, not programmatic success. The tradeoffs between carbon sources and sinks were considered issues of immense economic significance in the recent Kyoto climate negotiations. Better understanding of the carbon cycle will be of great value in the ongoing negotiations.

On the positive side, there are new and encouraging signs of focus and priority emerging from the NSTC/CENR structure and process.

Need for Program Balance

It can also be argued that there is currently an imbalance within the program among its major components: observing systems, data systems, and research and analysis. For instance, in the fiscal year 1996 USGCRP budget breakout, Our Changing Planet,²⁸ of the $1.83 billion allotted to the global change program, $1.19 billion (65%) was allocated to “Observing the Earth System” ($845 million) and “Managing and Archiving Data and Information ” ($343 million). Of the remainder, $434 million was allocated to “Understanding Global Change” (24%). As indicated above, this distribution of resources essentially continues in the fiscal year 1998 budget. It can be argued that the large investment required to develop and deploy the space observation component of USGCRP has comprised perhaps too large a fraction of the “focused” budget of the USGCRP. Nevertheless, the space missions designed to facilitate global change research, such as sea surface altimetry and scatterometry and the Upper Atmosphere Research Satellite, have been great successes. Moreover, after an 11-year hiatus, the capability to obtain ocean-color data has recently been restored with great scientific reward.

NASA's Earth Observing System (EOS) polar platforms—EOS AM-1, EOS PM-1, and EOS CHEM-1—were conceived as broadly scoped data-gathering systems. This foundation will be central for needed future missions, and it will set the baseline for a long-term, operational environmental monitoring program that must be built on the operational weather and ozone-observing system of NOAA, the US Department of Defense (DOD), and their international partners. However, while the EOS should begin to pay dividends with the scheduled 1998 launch of the AM-1 observatory followed by the late 2000 launch of the PM-1 mission, the initial focus of the USGCRP on EOS set a near-term timescale (and a cost) that made rapid response to scientific and technical challenges difficult.

The question of balance is further complicated by the realities of federal funding. Savings that might be obtained by trimming costs in NASA from space-based observations would be unlikely to flow within the agency to in-situ observational activities, let alone to the research and analysis (R&A) component (or even to other space-based missions). Still more unlikely is the transfer of such funds to other agencies within the USGCRP. These are political and institutional realities. Nevertheless, there remains the question of balance within the overall USGCRP observational system between space-based and in-situ systems. (In fiscal year 1996, only 11% of USGCRP observations were devoted to in-situ measurements.) Finally, although major breakthroughs have emerged from the research and analysis component of the national effort, it is just this part of the effort that continues to receive serious cuts within several agencies in the USGCRP.

Several lessons about Program balance can thus be extracted from the past 10 years. First, space-based observations are essential yet costly. We need to find ways of lowering their cost, while also making the space-based systems more budgetarily robust and flexible. We applaud NASA's Earth System Science Path-

finders and its rethinking of the EOS mission structure as steps in the right direction. Still another lesson is that in-situ observations are critical (e.g., the TOGA ocean buoy array for ENSO prediction); yet in-situ observational systems such as radiosonde and ozone networks continue to degrade around the world. We need to find ways to implement new in-situ observing systems while restoring and maintaining key existing systems. Finally, in recent years the scientific community has gone through a difficult experience: research and analysis budgets in critical areas have continued to decline, and science is simultaneously being asked for answers to increasingly difficult and important questions. We must find ways to reverse this declining trend (NSF's recently proposed fiscal year 1999 budget is a welcome change).

Need to Maintain Critical Observations

During the past 10 years, the value of critical combinations of models and observations has been repeatedly demonstrated in providing the nation and the world with critical information about specific issues of global environmental change.

The observing system that proved so valuable in the early detection of the present (1997–98) El Niño is a case in point. The research-based observing system and coupled atmosphere-ocean models developed under the auspices of the TOGA program to study ENSO phenomena made it possible as early as spring 1997 to detect and predict the present El Niño and its potential magnitude. Many social and economic systems are profoundly affected by weather events and climate patterns linked to ENSO; people in locations as distant as central Africa, southeast Asia, Australia, and North America are all benefiting from this scientific work, as agricultural, flood management, relief assistance, and market practices are adjusted.

Establishing an operational capability to maintain this initial ENSO observing system and training practitioners in the use of the data are large challenges, but there can no longer be any doubt that the investment has brought results of scientific interest as well as practical concern for natural resource management. This is an example of a crucial tenet of the Earth System Sciences Committee's strategy for studying global change: the institutionalization of critical measurement systems in an operational mode once their efficacy in documenting information valuable to policy makers is demonstrated in the course of a research program. This requirement will continue to be challenging for ENSO research; but more broadly the last 10 years have shown clearly that correctly transferring other key aspects of the observing program for USGCRP to operational programs will be very difficult.

This lesson also emerges clearly from negotiations on the polar platforms of NASA, NOAA, and DOD over the past 10 years. To date, the process is not a story of success for the USGCRP. For example, regarding the coordination among

the next generation of NOAA/DOD operational polar platforms and NASA EOS AM-1 and PM-1 satellites: if current plans proceed, there will be a significant gap between the conclusion of the flight of EOS PM-1 and the first NPOESS-1 (nominally planned for an afternoon crossing). ^e This gap will be significant because it will make coordination and calibration of the measurements taken by EOS PM-1 and NPOESS-1 extremely difficult.^f Beyond this specific issue and the continuing problem of adequately sequencing observations, there is a more general lesson to be learned: it is difficult for an operational program (e.g., NPOESS) to incorporate an adequate level of scientific advice, review, and essential oversight to ensure that the scientific needs of global change science will be addressed. This difficulty has been exacerbated until quite recently by NASA's distance from the NPOESS planning process; moreover, NPOESS itself is driven by two operational agencies (NOAA and DOD) with somewhat different demands on the data and data calibration and accuracy requirements, and it is understandable (but problematic) that global change issues are not high on the priority list.

The connectivity between EOS AM-1 and the future midmorning operational polar platform, EUMETSAT's METOP-2/3, is even more confused. This general issue brings to mind the additional difficulty of ensuring adequate coordination internationally, as possibilities are explored to transfer scientifically motivated observations to operational programs.

Other examples of problems are beginning to arise as research programs dependent on global observations of ocean, land surface, and atmospheric properties are concluding their intensive field campaigns. No provision is in place to make the necessary commitments for systematic acquisition of operational climate and global change in-situ data to continue the key time series started by these programs.

These are precisely the types of problems that the USGCRP was charged to resolve.

Need for Well-Calibrated Observations

During the past 10 years, we have been reminded again and again of the painful consequences of attempting to use inadequately calibrated observations to answer important questions about global environmental change. On a more positive note, great scientific advancements have been made when it is possible to use long-term, highly calibrated, rigorously maintained scientific observations. For example, precise measurements of atmospheric concentrations of carbon dioxide have yielded valuable information about the annual cycle of the biosphere and the distribution of carbon dioxide sources and sinks. Precise measurements

of CFCs have also enabled the tracing of atmospheric and oceanic circulations as well as improving the understanding of stratospheric ozone loss. Precise measurements of solar radiance have helped us to distinguish between natural and human influences on global mean temperature.

The general lesson here, then, is that high-quality data is an immensely powerful lever to obtain scientific insights on global change.

Need for a Focused Scientific Strategy

NRC reviews of the USGCRP over the past decade (see Bibliography), notably the intensive community-based review conducted at La Jolla in the summer of 1995,²⁹ have consistently emphasized the need for the Program to focus on critical scientific issues and the unresolved questions that are most relevant to pressing national policy issues. Pathways strongly reiterates this view. The nation and the world are beginning to make momentous decisions about development, technology, and the environment; at the same time, economic and political factors place severe constraints on budgets for research and infrastructure. A sharp focus on the truly essential investments in research and supporting infrastructure is thus more important than ever. A more sharply focused scientific strategy for the USGCRP is urgently required.

Charting and understanding the course of change in the Earth's physical, chemical, and biological systems, and their connections with human activities, are fundamental to the nation's welfare in the coming decades. Economic decisions, international negotiations, preservation of public health, and educational development demand this understanding. For example, without trusted knowledge about changes in the carbon and hydrologic cycles, ecological systems, temperature structure, storm systems, ultraviolet intensity, nutrient deposition, and oxidant patterns, defensible positions for international measures to protect the environment cannot be established and sustained.

Development of this urgently required knowledge will demand concerted efforts and continuing scientific leadership. As the world's leading scientific nation, the United States, working within the international community, must recognize the importance of providing scientific leadership in defining and diagnosing the changes in the state of the Earth system in the context of national needs and scientific interests. Strategic decisions on scientific goals, research programs, and supporting infrastructure are critical elements of this leadership, and it is the Committee's view that a new strategic approach is needed.

We thus present our findings and recommendations with the full sense of responsibility that accompanies the strong belief that the challenges posed to people by global environmental change will not go away. The challenges will not be legislated out of existence; they will be faced by our children's children, and they must be faced by us.

The CGCR is charged with providing scientific advice to the federal government on how the United States should execute global change research. Pathways addresses this task and the challenge of defining a new strategy for the USGCRP by identifying and considering the pivotal unanswered Scientific Questions in six fields: ecosystems, seasonal to interannual climate change, decadal to centennial climate change, atmospheric chemistry, paleoclimate, and the human dimensions of global change. For each field, the Committee discusses the character of the scientific problems; presents case studies associated with specific, relevant transitions in our scientific understanding of the Earth system; defines the primary unanswered Scientific Questions; critically reviews lessons learned in the course of achieving scientific transitions; and extracts from the analysis a set of Research Imperatives that, together with the corresponding, critical, unanswered Scientific Questions address fundamental needs to know in health, public policy, economics, international relations, and national leadership.

Observational priorities flow from the identified Research Imperatives and Scientific Questions, as do the required data and information systems required to manage these observations as well as some of the fundamental modeling issues that must be addressed to link the observations with the questions. Appendix C summarizes the Research Imperatives and primary Scientific Questions to which the United States research effort should be directed.

The Research Imperatives, which have been reviewed by the Committee in significant detail, provide the foundation for the findings and recommendations regarding actions needed to shape and implement the USGCRP over the coming decade. The Research Imperatives also set the direction and the metric to measure progress within the program.

Finding 1.1: Consideration of the identified Research Imperatives, case studies, and lessons extracted from two decades of research leads to the finding that vital improvements are possible in the execution of global change research. A large number of the most important advances in understanding the Earth system and in applying the findings to key policy questions have emerged from innovative combinations of individuals, observations, and modeling that attack specific questions. For example, the ENSO/TOGA program laid the groundwork for operational predictions to support natural resource decisions, and the stratospheric ozone research programs set the scientific foundation for the Montreal Protocol.

The fundamental scientific progress of the future will hinge on critical decisions about the character of the scientific program and the associated essential

observations. Resources have been most effectively utilized when applied in ways that strengthen the link between primary unanswered questions and the nation's intellectual resources, that improve the potential for technical innovations, that provide educational and public outreach opportunities, and that serve the vital information needs of decision makers.

Finding 1.2: Within each of the six topical themes identified in this report (see Appendix C) to further understanding of global change, the specific central scientific issues listed below must be confronted.

Finding 1.2a: Within Changes in the Biology and Biogeochemistry of Ecosystems, the following central scientific issues must be confronted:

• Understand the relationships between land-surface processes, including land-cover change, climate, and weather prediction.

• Understand the changing global biogeochemical cycles of carbon and nitrogen.

• Understand the responses of ecosystems to multiple stresses.

• Understand the relationship between changing biological diversity and ecosystem function.

Finding 1.2b: Within Changes in the Climate System on Seasonal to Interannual Timescales, the following central scientific issues must be confronted:

• Maintain and improve the capability to make ENSO.

• Define global seasonal to interannual variability, especially the global monsoon systems, and understand the extent to which it is predictable.

• Understand the roles of land-surface energy and water exchanges and their correct representation in models for seasonal to interannual prediction.

• Improve the ability to interpret the effects of large-scale climate variability on a local scale (downscale).

• Understand the seasonal to interannual factors that influence land-surface manifestations of the hydrological cycle such as floods, droughts, and other extreme weather events.

Finding 1.2c: Within Changes in the Climate System on Decadal to Century Timescales, the following central scientific issues must be confronted:

• Understand patterns in the climate system.

  • Natural Climate: Improve knowledge of decadal to century-scale natural climate patterns, their distributions in time and space, their optimal characterization, mechanistic controls, feedbacks, and sensitivities, including their interactions with, and responses to, anthropogenic climate change.

• Paleorecord: Extend the climate record back through data archeology and paleoclimate records for time series long enough to provide researchers a better database to analyze decadal to century-scale patterns. Specifically, achieve a better understanding of the nature and range of natural variability over these timescales.

• Long-Term Observational System: Ensure the existence of a long-term observing system for a more definitive observational foundation to evaluate decadal to century-scale variability and change. Ensure that the system includes observations of key state variables as well as external forcings.

  • Address the issues of those individual climate components whose resolution will most efficiently and significantly advance our understanding of decadal to century (dec-cen) climate variability.

Finding 1.2d: Within Changes in the Chemistry of the Atmosphere, the following central scientific issues must be confronted:

• Define and predict secular trends in the intensity of ultraviolet exposure the Earth receives by documenting the concentrations and distributions of stratospheric ozone and the key chemical species that control its catalytic destruction and by elucidating the coupling between chemistry, dynamics, and radiation in the stratosphere and upper troposphere.

• Determine the fluxes of greenhouse gases into and out of the Earth 's systems and the mechanisms responsible for the exchange and distribution between and within those systems.

• Develop the observational and computational tools and strategies that policy makers need to effectively manage ozone pollution; elucidate the processes that control and the relationships that exist among ozone precursor species, tropospheric ozone, and the oxidizing capacity of the atmosphere.

• Improve atmospheric models to better represent current atmospheric oxidants and predict the atmosphere's response to future levels of pollutants.

• Document the chemical and physical properties of atmospheric aerosols; elucidate the chemical and physical processes that determine the size, concentration, and chemical characteristics of atmospheric aerosols.

• Document the rates of chemical exchange between the atmosphere and ecosystems of critical economic and environmental import; elucidate the extent to which interactions between the atmosphere and biosphere are influenced by changing concentrations and depositions of harmful and beneficial compounds.

Finding 1.2e: Within Paleoclimate, the following central scientific issues must be confronted:

• Document how the global climate and Earth's environment have changed in the past and determine the factors that caused these changes. Explore how this knowledge can be applied to understand future climate and environmental change.

• Document how the activities of humans have affected the global environment and climate and determine how these effects can be differentiated from natural variability. Describe what constitutes the natural environment prior to human intervention.

• Explore the question of what are the natural limits of the global environment and determine how changes in the boundary conditions for this natural environment are manifested.

• Document the important forcing factors that are and will control climate change on societal timescales (season to century). Determine what were the causes of the rapid climate change events and rapid transitions in climate state.

Finding 1.2f: Within Human Dimensions of Global Environmental Change, the following central scientific issues must be confronted:

• Understand the major human causes of changes in the global environment and how they vary over time, across space, and between economic sectors and social groups.

• Determine the human consequences of global environmental change on key life-support systems, such as water, health, energy, natural ecosystems, and agriculture, and determine the impacts on economic and social systems.

• Develop a scientific foundation for evaluating the potential human responses to global change, their effectiveness and cost, and the basis for deciding among the range of options.

• Understand the underlying social processes or driving forces behind the human relationship to the global environment, such as human attitudes and behavior, population dynamics, institutions, and economic and technological transformations.

An additional finding flows from the full report and Recommendation 1.

Finding 1.3: In spite of the initial efforts to encompass a broad view of the Earth system, certain critical research areas continue to suffer either because their relationship with global change research was not clearly articulated initially or because they cross over disciplinary boundaries of environmental science. For example, the important issue of biodiversity is not adequately addressed by the USGCRP. Biodiversity research is often quite germane to global change research (and vice versa)—for example, see Finding 1.2a—and there are important interactions between global change and biodiversity loss, but biodiversity research also has major components and activities that are beyond the scope of global change research. The US science community has been unable to resolve related boundary issues. Because the CGCR is recommending a sharpening of focus for the USGCRP, the issue of addressing more fully the scientific issues posed by biodiversity is more likely to be left unresolved unless there is a deliberate effort by the USGCRP agencies and the NRC to help resolve this problem. The emergence of the DIVERSITAS^g program demonstrates that it is beginning to be addressed better internationally.

Two common, linked themes emerge clearly from the identified Research Imperatives, Scientific Questions, and the associated observations (elaborated in Appendix C and described in the full report):

• The water and carbon cycles

• Issues of climate prediction, including the role of and impacts on human and generational (30 year) timescales and at spatial scales useful for critical public and private policy decisions.

The scientific strength of the case linking release of greenhouse gases to climate change is central to any considered action and must be commensurate with the economic impact of any proposed solution. Acceptance of the evidence by policy makers and their constituent communities is a key component of public policy negotiations. This acceptance will emerge not merely from refinement of assessment models but rather from carefully executed observations of the physical, chemical, and biological variables that track the actual state of the Earth. It is the accumulation of evidence from a body of studies, adhering to strict scientific standards and focus, that will provide a basis for decisions, not the results of a single study.

Knowledge of the global sources and sinks of carbon that are associated with changes in atmospheric concentrations of carbon dioxide, methane, and carbon monoxide is part of the foundation for understanding the physical, chemical, and biological processes that control our surroundings and for understanding the fractional impact of any industrial or agricultural input to that natural system.h

Similarly, water is at the heart of both the causes and the effects of climate change. It is essential to establish rates of and possible changes in precipitation, evapotranspiration, and cloud water content (both liquid and ice). Additionally, better time series measurements are needed for water runoff, river flow, and, most importantly, the quantities of water involved in various human uses. This crosscutting initiative can clearly build upon the progress made by the Global Energy and Water Experiment (GEWEX) in the World Climate Research Program and the Biospheric Aspects of the Hydrological Cycle (BAHC) project of the International Geosphere-Biosphere Program.

Elucidating the climate system and possible anthropogenic changes, in addition to natural variability, are paramount goals in these studies. A satisfactory demonstration of secular trends in the Earth 's climate system, for example, requires analysis at the forefront of science and statistical analysis. Model predictions have been available for decades, but a clear demonstration of their validity, a demonstration that will convince a reasoned critic on cross examination, is not yet available. This is not in itself either a statement of failure or a significant surprise. Rather, it is a measure of the intellectual depth of the problem and the need for carefully orchestrated, long-term observations.

Finding 2.1: Key crosscutting scientific themes that emerge from the total set of Research Imperatives and Scientific Questions must be addressed in an integrated fashion.

Finding 2.2:

• International negotiations will hinge on the strength of the evidence defining global scale sources and sinks of carbon. Innovative approaches exist that can provide vital knowledge to define current carbon sources and sinks.

• Water is at the heart of climate change and the impacts of climate variability. Any assessment of climate change, its causes and impacts, must be based on significantly better observations of the water cycle.

• Observing, documenting, and understanding climate change is a central responsibility of the US effort in global change research. Predictive capabilities must be developed on temporal and spatial scales particularly relevant to the coming generation of citizens. Scientific focus, continuity,

  h	Carbon monoxide is not a greenhouse gas, but it is involved with the chemistry of other greenhouse gases (e.g., carbon dioxide and methane).

and insight are critical ingredients in this pursuit. An innovative combination of observations is required on all timescales: seasonal to interannual and decadal to centennial.

The USGCRP must develop an approach that satisfies a number of critical objectives:

• Improves the ability to establish accurate time series of spatially resolved flux measurements of carbon species and their isotopes and associated observations of molecular oxygen

• Clarifies the distribution and fates of water

• Establishes the spatial and temporal distribution of the phases of water in the middle-upper troposphere

• Defines climate change on temporal and spatial scales relevant to current and emerging issues of public policy

• Provides the capability to resolve sharp nonlinearities within the Earth system that are triggered by chemical composition changes, which in turn lead to phase changes that markedly affect the transport of infrared radiation.

The Research Imperatives help to identify preliminary emphases for required observations and data systems and to focus the needed calculations and models; the Scientific Questions provide the specificity required to establish what must be done to advance our understanding. The required observations (whether relating to long-term trends in radiance to space, fluxes of carbon-containing molecules into and out of a particular ecosystem, chemical and isotopic composition in ice core samples, depth profiles of temperature and salinity, rate-limiting free radical concentrations as a function of nitrogen loading, or other phenomena) are out-

lined, where possible, with respect to accuracy, spatial and temporal resolution, required simultaneous measurements, and other defining characteristics so that each measurement ensemble is formulated to answer a specific primary Scientific Question.

The importance of accuracy, continuity, calibration, documentation, and technological innovation in observations for long-term trend analysis of global change cannot be overemphasized. A central tenet of the Committee's analysis is the necessity for the continuity of key global change observations.ⁱFor example, with regard to the fundamental forcing parameters of global change, such as solar radiation and atmospheric carbon dioxide concentration, and response parameters, such as surface temperature and global cloudiness, discontinuities in the climate record resulting from instrument changes or drift have led to questions about the very nature of global change. Instrument or technology changes per se are not the problem; the problem is inadequate cross-calibration between instruments, and this inadequacy usually results from the absence of commitment to observational continuity. The Research Imperatives identified in this report express guiding considerations for the USGCRP to fulfill its responsibility for observing, documenting, and understanding global environmental change.

The critical nature of high-quality observations to the scientific and public policy issues posed by global environmental change places demands and constraints on whatever path a USGCRP observational strategy attempts to chart; however, a specific, well-considered, and realistic strategy, including costs and schedule, for obtaining the observations of past, present, and future expressions of global environmental change is essential. The strategy will need an effective institutional mechanism for implementation. As an example, no agency currently has the responsibility for carrying out or coordinating a comprehensive program of climate observations.

There are many different scientific demands for observations for exploratory surveys, hypothesis testing in coordinated process studies, repetitive analysis-forecast cycle research, documentation of long-term changes, calibration and validation of measurements, and applications or modifications of measurements that may be used primarily for purposes other than global change research. These different demands or applications must be taken into consideration in developing a coherent observational strategy.

Operational demands are uniquely linked to long-term measurements and therefore are vital for obtaining them. In particular, the satellite and in situ measurements taken as part of the weather-observing system are critical to the future of the climate record. Development of the next generation of weather satellites (e.g., the National Polar-Orbiting Operational Environmental Satellite System, NPOESS) should be undertaken with the climate record and other research on relevant global change Scientific Questions clearly in mind.

Documentation of decadal and longer term change raises other basic issues of program management and decision structure. Because adequate characterization of higher frequency variability is fundamental to this documentation, both to avoid aliasing and to help in attributing causes, there is major overlap between long-term observations and measurements that are necessary on the daily and interannual timescales. However, as the timescale of the phenomenon studied becomes longer, two other considerations become increasingly important for adequate management of the research enterprise.

First, the need for comparability of measurements made at different times and places requires that high priority be given to thorough instrument calibration and measurement system validation, including the inevitable changes in technologies and observing networks. Because action is required now, but all the specific Scientific Questions may not come into focus for many years, it is necessary to invoke the concept of stewardship to justify this effort. Stewardship involves doing what is reasonable and prudent to safeguard the interests of future generations, who are not able to argue their case for the data and information.

Second, to put even reliably observed interdecadal changes in context, it is necessary to invoke records of much longer duration than available based on modern instrumentation. Thus, the strategy must include the systematic search for, and recovery and exploitation of, naturally existing proxies for such instrumentation, proxies that reveal the past history over hundreds and thousands of years with adequate fidelity and temporal resolution. This activity would appear to have little relationship to what is conventionally known as an observing system. Moreover, both calibration of such proxy records in terms of modern instrumental measurements and execution of process studies aimed at their interpretation are less glamorous tasks than launching satellites to observe fine details over the next decade or so. Nevertheless, these less glamorous activities may yield much more useable information for the foreseeable future about the natural processes leading to environmental fluctuations on such timescales and hence about the modifications induced by human activities.

A coherent observational strategy is needed that builds on the identified Research Imperatives and Scientific Questions and on available national and international space and in-situ networks. The USGCRP must find a mechanism to resolve the agency boundary issues that will surely arise in developing and especially in implementing a coherent observational strategy. The United States and its international partners must find a way to deal effectively with the international dimensions of an overall observational system. In sum, what is needed is not a vast new program but rather attention to coordinating, simplifying, and focusing current and planned observing systems. This work requires the sustained attention of the scientific community and the farsightedness of government to ensure the survival of key observational records. A particular challenge will be the in-situ systems.

Finding 3.1: Although extensive planning has been done for space-based systems to observe global climate, the oceans, and the land, a comprehensive space-based system does not yet exist in practice. It is a promise that remains unfulfilled. Moreover, it is not clear that current planning activities will lead to such a system. Central issues about which nation (or nations) will provide which observations and for how long and at what spatial and temporal scales (and with what assurance) remain unresolved. The situation for in-situ observations across the full global environmental change agenda is in far worse shape.

Finding 3.2: The connectivity of NOAA's NPOESS program with NASA's Earth Observing System in Mission to Planet Earth is an important and not yet adequately resolved issue. The adequacy of the NPOESS measurements to meet the demands of global change research remains in question. In addition, it is essential to maintain those stations of the existing in-situ weather observation network of the United States and around the world that carry the climate record from past decades. The current and future state of this system is unclear.

Although we recognize the danger in recommending another study or planning exercise, a path to a more realizable, logical, focused, and robust observing system must be found. The USGCRP must adopt multiple observational approaches, recognizing that no single approach can guarantee continuity and accuracy of measurements and that independent checks are necessary to obtain verifiable results.

Given the constraints on the budget system and the needs of the research community for observations from space, the strategy appears to involve three components:

• Within NASA, build focused, less costly missions on the solid and broad foundation set by EOS.

• Within NOAA, build scientifically sound observational missions for monitoring global change on the foundation set by EOS; these missions must meet NOAA's operational requirements.

• Within the USGCRP, increase the funding for in-situ observational programs and the necessary research and analysis links necessary for the related essential science.

The first component should be possible within the scope of current budget projections. The second component may require additional funds for NOAA; it may require modification of NOAA's mission (e.g., a strong commitment by NOAA to address global environmental change as part of its mission), and it definitely requires significantly improved coordination between NOAA and NASA. The third component requires both new funds, which have begun to appear in the proposed fiscal year 1999 budget, and sharper focus in using existing funds. With regard to the second component above, it is crucial to recognize that, even if NOAA were to assume prime responsibility for the USGCRP space-based, monitoring program, NASA would continue to have significant data-processing responsibilities including reanalyses. Finally, the issue of the DOD role and influence on the observational, space-based monitoring program must be addressed by USGCRP.

Innovation is essential for scientific progress in this global change research. Many needs illuminate the importance of innovation, foresight, and testability in this field:

• Obtaining simultaneous, high-resolution observations with high sensitivity of the sea surface and the marine boundary layer

• Determining fluxes of carbon species into and out of broad categories of ecosystems

• Establishing patterns of land use and the state of vegetation

• Observing the vertical profiles of temperature, salinity, velocity, and tracer concentrations in the oceans

• Establishing the distribution of water in the atmosphere and the fluxes of water between the Earth's surface and the Earth's atmosphere

• Obtaining isotopic composition of water in the middle/upper troposphere

• Determining systematically the concentrations and concentration derivatives of catalytically active free radicals at altitudes from the sea surface to the middle stratosphere

• Obtaining observations along Lagrangian trajectories to dissect aerosol formation processes.

There is also a fundamental problem in global change observations that can be attacked only by technical innovation. The ocean-atmosphere-biosphere is seriously undersampled—mechanistically, spatially, and temporally.

Finding 4: The capability, availability, cost, and character of observational platforms are critical considerations in global change research strategies. Observational platforms are the foundation of the nation's research efforts, and the design of these platforms can profit significantly from the lessons learned in carrying out global change research to date. Consideration of these lessons demonstrates that the successful execution of global change research is closely tied to technical innovation. Investment in observational platforms to date has been focused on a small number of large satellites, a limited number of marginally funded aircraft, a small number of ocean buoy systems, and a sparse network of ground-based efforts. This balance among the space-based, airborne, and ground-based observations does not reflect the spectrum of requirements that the Research Imperatives demand. Indeed, space-based observations and their associated data-management systems dominate the resources of the USGCRP, a trend that impinges on both the research and analysis support and the in-situ observational networks.

The issue of data systems and the design of those systems closely tied to the character of the observational strategy and the associated theoretical and modeling effort used to address the important questions. A key common component of major scientific advances has been the focus of responsibility: a specific principal investigator (or close collaboration of coinvestigators) must bear the end-to-end responsibility that connects the posing of a scientific question to the execution of an observational strategy with associated theoretical analysis and through to the publication of scientific conclusions in the refereed scientific literature. A plan in

which committees and/or agencies are assigned responsibility for data quality and distribution in a manner that breaks the end-to-end responsibility of the principle investigator is almost invariably critically flawed. The scientific method depends on a strategic combination of observations, selected from an array of possible observables, that can dissect a problem to the satisfaction of peer critics. This achievement demands specific choices, and it demands focused responsibility. The NRC Committee on Data Management and Computation has already shown that effective data systems require continuous and widespread involvement of the science team.

Connected with the fundamental role of the investigator in assuring appropriateness and quality of observations is the rapidly advancing state of information systems, which can allow distribution of activities in time and space while preserving the essential responsibility of the scientific investigator. It is important, in considering the scientist and the information system, to consider the nature of the future of this interaction.

The evolution of information systems will likely be characterized by rapid, dramatic shifts, as much as by any smooth, “predictable ” process. In an industry that shows a quadrupling of capability every 3 years, there are no stationary solutions. Stability and success can be attained only by development of a solid, well-grounded information model that describes how the pieces and subsystems, including as they develop, are related to each other. This model should be based on science that incorporates a database-driven approach; the technical implementation can then be more flexible and take advantage of technological advances in a more rational manner.

To date, the concentration has been on processing and storage, but the network infrastructure as well as the software is undergoing fundamental changes. Although the scientific community has logically paid attention mainly to such government and academic backbones as vBNS and Internet-2, the more important shift is occurring in the widespread distribution of high bandwidth (1-10 Mbps) to the home. This shift has an implication for the USGCRP. First, more home users will likely be searching NASA, NOAA, and other global change archives for interesting or educational material. The networking capabilities of these more informal users will be competing with scientists for access to the archives. Second, these users will likely demand different types of products than scientists. This probable situation needs to be recognized.

The NSF Knowledge and Distributed Intelligence solicitation in the fiscal year 1999 budget is an example of government-encouraged partnerships with the private sector that may accelerate this trend. Just as the Internet has changed the model of how we conduct research, so these new distribution channels will change our model again.

Ultimately, the USGCRP is about information. Information must flow within the Program and also to the broad community of users. The subject of the Program's research demands that information flow effectively to the public at

large as well as to researchers. This is an important issue, and it should not be ignored by either the community of scientists engaged in global change research or by the agencies that support this research with public funds.

Finding 5: Data systems must be agile and responsive to technology developments and to emerging techniques for data handling, analysis, and transfer. Data systems must also maintain scientific discipline and focused responsibility, so that the link between scientific question and clear scientific conclusion is not broken. An appropriate system is one that charges the government with initiallevel processing and long-term archiving and that charges the scientific community with producing the scientific products by the most effective means possible.

As suggested in the last finding, it is likely that the government will continue to provide the primary long-term archive for space and Earth science data, but it must also maintain the capability to enable long-term reprocessing of these time series; archiving must not continue to be a burial ground for data. With more rapid distribution channels and more powerful archive and processing systems at the fringes, perhaps one part of the government's role is to provide an on-line repository of data recipes rather than fully processed data sets. This service would enable more customized processing with the government serving as the warehouse for raw materials and generating specific products on demand. Finally, changes in technology will allow and force us to rethink our strategy often; any strategy must accommodate and encourage this eventuality.

As mandated by its implementing legislation, the USGCRP seeks to provide useful information to the policy process. A direct implication of this responsibility is that the information must be scientifically credible, that it be of genuine interest and value, and that, to the greatest extent possible, it provide lead time for policy action. The last requirement implies provision of some prognostic infor-

mation. This requirement does not necessarily entail a “prediction,” but it does raise the same concerns as any prediction or predictive process. These concerns revolve around general, and not necessarily scientific, issues such as usefulness, trustworthiness, and credibility of the information. In general, a model or a set of models will often be at the center of the predictive process.

Finding 6: The policy issues that confront global change research, like the Scientific Questions, are serious, particularly with regard to their impact on humans. These issues will rely on models of exceedingly complex behaviors over a significant range of scales in space and time. Significant challenges face the scientific community in the form of many and various modeling issues, from initialization to validation. Important, unsolved, difficult problems remain for formulating useful prognostic models over a range of topics in human-dimensions research. Advances in developing and most importantly in testing and evaluating models are needed. The United States is no longer in the lead in this critical field.

The fact that the United States is no longer in the lead in applying global models is not purely a statement of criticism. Strong scientific work, particularly in the area of modeling, has been advancing around the world. This is to be applauded. Global change research, particularly in the area of prognostic activities, requires a full suite of models to adequately bracket the complex problems that USGCRP seeks to address. Thus, advances in modeling capabilities in other parts of the world are of significant benefit to the USGCRP. Testing adequately complex models is very computing intensive, and if computing resources are not adequate and available then there is clearly the danger that the dynamical aspects of models will not be sufficiently understood and hence that the models will be misapplied. Currently, the potential exists that the advanced models built in the United States cannot (or will not) be adequately tested and properly applied to key problems, such as national and regional expressions of transient climate variability and change because of a lack of available computing resources. The United States must apply greater resources, particularly (but not exclusively) in the area of advanced computing machines. National boundaries should not influence where machines are purchased.

Models must be tested and evaluated with observations. This means that adequate observations and advanced computing resources must be available to adequately evaluate models and their potential utility for the public policy process. Consequently, there must be a greater commitment to advanced computing resources, as well as human resources, by the USGCRP to ensure that global modeling is achieved at spatial and temporal scales appropriate to the needs of the policy community and the private sector.

As the USGCRP enters this second critical decade of its existence, the scientific challenges it faces are heightened by the need to understand and foreshadow the regional, as well as other, impacts of global environmental changes. The causes of global change are now also more complex, the need to understand the effects of multiple stresses are more apparent, and the likelihood of realizing significant near-term global reductions that would lead to stabilization of the forcing terms (such as greenhouse gas concentrations in the atmosphere) before a doubling of the radiative effects are more remote. In short, the need for useful prognostic information will only increase in the future. In view of these considerations, the current circumstances within the USGCRP, and the current status of modeling and available computing resources to the global change scientific community, there must be a considerably expanded commitment of resources to modeling, particularly at the temporal and spatial scales needed by the policy community.

Goody, R. 1982. Global Change: Impacts on Habitability. Report by the Executive Committee of a Workshop held at Woods Hole, Massachusetts, June 16-21, 1982. JPL D-95. National Aeronautics and Space Administration, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, 15p.

Earth System Sciences Committee (ESSC), 1986. Earth System Science. Overview. NASA Advisory Council, National Aeronautics and Space Administration Washington, D.C., 48p.

Earth System Sciences Committee (ESSC), 1988. Earth System Science. A Closer View. NASA Advisory Council, National Aeronautics and Space Administration Washington, D.C., 208p.

Fein, J. S., P. L. Stephens, and K. S. Loughran. 1983. The Global Atmospheric Research Program: 1979-1982. Reviews of Geophysical and Space Physics. 21: 1076-1096.

International Council of Scientific Unions (ICSU), 1996. Understanding Our Planet. ICSU Press, Paris, France, 48p.

Intergovernmental Panel on Climate Change (IPCC), 1995. Climate Change 1995: IPCC Second Assessment Report. Cambridge University Press, 448p.

Massachusetts Institute of Technology (MIT), 1970. Man's Impact on the Global Environment. Report of the Study of Critical Environmental Problems (SCEP). MIT Press, 319p.

Massachusetts Institute of Technology (MIT), 1971. Inadvertent Climate Modification. Report of the Study of Man's Impact on Climate (SMIC). MIT Press, 308p.

National Research Council (NRC), 1966. The Feasibility of a Global Observation and Analysis Experiment. Committee on Atmospheric Sciences. National Academy Press, 172p.

National Research Council (NRC), 1979. Carbon Dioxide and Climate: A Scientific Assessment. Ad Hoc Study Group on Carbon Dioxide and Climate. National Academy Press, 22p.

National Research Council (NRC), 1982a. Energy and Climate. Geophysics Study Committee. National Academy Press, 158p.

National Research Council (NRC), 1982b. Carbon Dioxide and Climate: A Second Assessment. CO₂/Climate Review Panel. National Academy Press, 72p.

National Research Council (NRC), 1982c. Causes and Effects of Stratospheric Ozone Reduction: An Update. Board on Environmental Studies and Toxicology. National Academy Press, 339p.

National Research Council (NRC), 1983a. Toward an International Geosphere-Biosphere Program. National Academy Press, 81p.

National Research Council (NRC), 1983b. Changing Climate. Carbon Dioxide Assessment Committee, Board on Atmospheric Sciences and Climate. National Academy Press, 486p.

National Research Council (NRC), 1983c. El Niño and the Southern Oscillation: A Scientific Plan. Board on Atmospheric Sciences and Climate. National Academy Press, 72p.

National Research Council (NRC), 1986. Global Change in the Geosphere-Biosphere. Initial Priorities for an IGBP. U.S. Committee for an International Geosphere-Biosphere Program. National Academy Press, 91p.

National Research Council (NRC), 1991. Policy Implications of Greenhouse Warming. Committee on Science, Engineering, and Public Policy. National Academy Press, 944p.

National Research Council (NRC), 1995. A Review of the U.S. Global Change Research Program and NASA's Mission to Planet Earth/Earth Observing System (La Jolla Report). Committee on Global Change Research and Board on Sustainable Development. National Academy Press, 96p.

National Research Council (NRC), 1996. A Review of the U.S. Global Change Research Program. Committee on Global Change Research. Letter report, 20p. [Post-La Jolla Report.]

Revelle, R., and H. E. Suess. 1957. Carbon dioxide exchange between atmosphere and ocean and the question of an increase of atmospheric CO₂ during the past decades. Tellus 9: 18.

United Nations Environment Program (UNEP), 1994. Scientific Assessment of Ozone Depletion 1994. United Nations Environment Program, Geneva, 578p.

U.S. Department of Energy (DOE), 1977. Workshop on the Global Effects of Carbon Dioxide from Fossil Fuel. William P. Elliot and Lester Machta, eds. Held March 7-11, Miami Beach, FL.

U.S. Department of Energy (DOE), 1980. Workshop on Environmental and Societal Consequences of a Possible CO₂—Induced Climate Change. Conducted by the American Association for the Advancement of Science on April 2-6, 1979, Annapolis, MD.

USGCRP, 1992. Our Changing Planet: the FY 1993 U.S. Global Change Research Program. A Supplement to the President's Fiscal Year 1993 Budget. U.S. Global Change Research Program Office, 79p.

USGCRP, 1995. Our Changing Planet: the FY 1996 U.S. Global Change Research Program. A Supplement to the President's Fiscal Year 1996 Budget. U.S. Global Change Research Program Office, 151p.

USGCRP, 1997. Our Changing Planet: the FY 1998 U.S. Global Change Research Program. A Supplement to the President's Fiscal Year 1998 Budget. U.S. Global Change Research Program Office, 118p.

World Meteorological Organization (WMO), 1979. Proceedings of the World Climate Conference. World Meteorological Organization, Geneva.

World Meteorological Organization (WMO), 1984. Report of the Study Conference on Sensitivity of Ecosystems and Society to Climate Change, Villach, Austria, Sept. 1983. WCP Publication 83. World Meteorological Organization, Geneva.

World Meteorological Organization (WMO), 1986. Report of the International Conference on the Assessment of the Role of Carbon Dioxide and of Other Greenhouse Gases in Climate Variations and Associated Impacts, Villach, Austria. World Meteorological Organizations, Geneva. Publication 661, 78p.

World Meteorological Organization (WMO), 1990. Proceedings of the Second World Climate Conference. World Meteorological Organization, Geneva.