physical state of the upper ocean will be systematically measured and assimilated in near real-time.

Completion of the planned-for 3000 floats will be important to the achievement of several scientific advances:

(1) improvements in the predictability of season-to-decadal climate variability; (2) quantitative description of the evolving state of the upper ocean and the patterns of ocean climate variability; and

(3) provision of data inputs for the initialization and evaluation of ocean and coupled ocean-atmosphere forecast models.

The Argo array is part of the Global Climate Observing System/Global Ocean Observing System (GCOS/GOOS) and part of the Climate Variability and Predictability Experiment (CLIVAR) and the Global Ocean Data Assimilation Experiment (GODAE).

In addition to the Argo observing system mentioned above, there are specific ocean observations that are important to a better understanding the role of the oceans in the climate system and to better characterize the key climate-system processes that currently limit prediction and hence decision-making associated with climate change. A few of the most pressing ones are noted below. In addition, to address the Committee's question about how to gain the highest degree of confidence in answering the question about how climate has changed, the answer below also describes the corresponding needs associated with the air-land-surface interface.

• Clarifying what are the "choke points" of the lame-scale oceanic circulation. Research Foci: Models and paleo-climate studies indicate that the large-scale, density-driven ("thermohaline") oceanic circulation plays an important role in long-term atmospheric climate. In addition to periodic variability, this circulation may play an important role in abrupt climate change, a possibility that has attracted considerable attention. Since the thermohaline circulation is global, a cost-effective and efficient method of monitoring this component of oceanic circulation is needed. Continuous measurements of water mass properties (e.g., temperature, salinity, oxygen, etc.) and transport at a limited number of locations can provide these data. Specifically, flow through the passages south of Africa and South America represent such chokepoints. In addition, flows out of the Arctic and Mediterranean are also important. Additional information about narrow-channel flows (e.g., between Iceland and Greenland) and boundary (i.e., near land) flows would substantially improve the understanding of this circulation and its changes. Payoffs: Data and analyses at these choke-points will provide (i) information on the intensity of the thermohaline oceanic circulation and early warning signs for any abrupt climate change and (ii) improved predictive understanding of the role of the thermohaline circulation in the climate system. Hence, speculation would be narrowed, and quantitative simulations of potential change (of lack of change) of this major climate feature would provide important information relevant to society.

• Improved observations of climatically important gas exchanges between atmosphere and ocean. Research Foci: Surface atmospheric data (wind, heat flux, etc.) are regularly collected from Voluntary Observing Ships (VOS), fixed buoys, and surface drifters, for example. Techniques for measuring gas exchange via these types of platforms are also needed. For example, CO2 exchange between ocean and atmosphere is an important component of the global cycling of carbon between oceans and atmosphere. In addition, detailed water-column measurements made in the past should be continually revisited (on a five to ten-year cycle) to determine the total water column inventory of heat, salt, and climatically important gases. Previous studies have shown that the ocean is warming, but the data used in these studies will not be continuously available in the future. Only some ten transects will be needed globally to provide this information. Payoffs: These data will (i) help ensure that climate signals can be distinguished from instrument signals, (ii) provide continuous calibration/validation observations for satellite-based estimates of gas exchange processes; (iii) help provide the information needed to attribute the cause of climate signals; and (iv) help determine the ocean's role in climate change and to attribute this change to natural and/or anthropogenic causes. Characterizing land-surface/biosphere/atmosphere interactions. Research foci: Land-surface changes provide important feedbacks in the climate system since climate changes influence the state of the land surface (e.g., soil moisture, reflectivity, and vegetation cover). Requiring better characterization are,

71-800 D-01--3

for example, large-scale deforestation and reforestation, including their impact on the hydrological (water) cycles. Payoff: Being able to incorporate such feedbacks better into climate models will allow the past rich historical and the growing paleo data base to test effectiveness of the needed sub-global predictive skills.

• Understanding abrupt climate changes: When and why? Research foci: Paleoclimate data reveal that relatively abrupt and sustained climatic shifts have occurred in the past. The formation of the Sahara Desert about 5,500 years ago is an example of such an abrupt change (perhaps from a non-linear change in land cover, which relates to the preceding bullet). Focusing paleoclimatic and diagnostic modeling on such events could better identify threshold mechanisms, such as those that could be related to abrupt shifts in the oceanic circulation (as noted in the first bullet in the above section). Payoffs: It would be enormously important and relevant to decision-making to be able to assess quantitatively the likelihood of an abrupt climate shift during the coming century, since current climate models are simulating a rate of warming of global temperatures without precedent during at least the last 10,000 years.

• Characterizing water vapor and clouds. Research Foci: The water-vapor feedback process amplifies (by a factor of two) the greenhouse role of all other greenhouse gases. Further, clouds are a key factor in the albedo (i.e., reflectivity) of the planet, as well as many "feedback" processes. Focused remotesensing and in-situ studies with new techniques could improve this understanding substantially, when combined with diagnostic interpretation of the challengingly-large spatial and temporal variability. Payoffs: Understanding these processes would address what is probably the largest modeling deficiency that contributes to the wide uncertainty range of simulated temperature increases that would be expected for a given future emission scenario. This research would address the fundamental need for the "scientific" uncertainty associated with any particular option (scenario) to be less than the difference between the outcomes of various different options (i.e., being able to demonstrate that there is a real benefit).

Question: (Asked to all witnesses) The recent Kyoto negotiations at the Hague were stymied in large part due to disagreements over how efficient plants are at tying up carbon from the atmosphere. Are the current carbon cycle programs sufficient to obtain the understanding we need so that we can make appropriate policy decisions? Answer:

Indeed, discussions focusing on altering greenhouse gas emissions are confronted with information gaps that limit the ability to estimate better the future atmospheric concentrations/climate-forcing of atmospheric constituents. Carbon dioxide is major greenhouse gas, and the first point below addresses associated needs for scientific understanding on that topic. My answer also includes important information regarding "other" greenhouse gases that could potentially play a role in facilitating the broad discussions that occur in policy negotiations. Two of the most-immediate policy-relevant research needs are the following:

• Better parameterization of the carbon cycle. Research foci: While the source of human-influenced carbon emissions is relatively well quantified, the fate and time history of carbon emissions are not as well known. Better information on the annual/decadal terrestrial processes involved (e.g., biospheric uptake/release) is required for defensible options for carbon management. Similarly, characterization of processes involved in the large carbon-uptake, longterm storage role of the oceans is a key to better establishing the fraction of emitted carbon that remains in the atmosphere in the decadal/century time frame. Payoffs: Having a better quantitative model of the carbon cycle, including how climate change will influence it, will yield more credible simulations of future CO2 abundance (hence, radiative forcing) for specified choices on emissions ("What change in radiative forcing results from what emission changes?").

• Quantify radiative roles, trends, and variations of the shorter-lived greenhouse gases and aerosols. Research Foci: Ozone in the lower atmosphere (troposphere) and aerosols (fine airborne particles) play unique (but poorly quantified) roles in climate change. Both cause regional radiative forcing, but of opposite sign (warming for ozone, cooling for sulfur-containing aerosols, and warming for carbon-containing aerosols), which is very unlike the global distribution of carbon dioxide. Further, these constituents have short atmos

pheric residence times, and therefore afford the means for changing climaterelated driving factors in the near term (unlike the slow decadal response associated with CO2). Further, tropospheric ozone and aerosols are associated with other environmental issues (e.g., poor air quality). Major information gaps could be addressed by process and modeling studies that better link emissions to global distributions and to radiative forcing patterns, both past and future. Payoffs: In short, to improve the now-weak quantitative knowledge of the shorter-lived constituents would (i) open additional quantified options for changing climate-related driving factors, (ii) provide information on how choices associated with one issue would influence another issue, and (iii) substantially improve the level of confidence in model predictions of, say, temperature increase over this century.

Question: (Asked to all witnesses) The USGCRP is a collaborative multi-agency initiative. How can global climate research be strengthened given the dispersed nature of the initiative? Does the USGCRP umbrella of agencies have a coordinated approach for prioritizing, from a national perspective, their climate modeling research and assessment efforts?

Answer:

The U.S. Global Change Research Program is a multi-Agency effort, with the inherent advantages and disadvantages of such diversity. The Program is currently establishing its pathways and procedures for its second decade by developing a tenyear plan that focuses on the highest priority research issues.

Question: (Asked to all witnesses) Are human and fiscal resources allocated effectively to address the above mentioned priorities? Are students being trained to fill either the scientific research positions or the niches of computational science and software engineering required for a successful high-end climate computing capability? Answer:

Current budgets effectively allocate resources for climate change research priorities. The President's Global Climate Change Initiative will set priorities for additional investments in climate change research. The initiative is planned to fully fund high-priority areas for climate change science over the next five years. Regarding students, the academic colleagues that testified with me are far better informed about the educational needs and how well they are being met.

Question: (Directed to Dr. Albritton) Is it possible that the warming we're seeing is part of some natural fluctuations, some kind of “noise” if you will, in the system? Answer:

In Chapter 12 ("Detection of Climate Change and Attribution of Causes") of the IPCC, the researchers assessed this likelihood raised in the question above. The assessment compared the observed global-average surface temperature changes to those simulated by climate models for three Cases: (a) natural variation, (b) anthropogenic climate-change forcing, and (c) the combination of natural variation and anthropogenic forcing. As shown in Figure 4 of the IPCC Summary For Policymakers, the best match was Case (c). The mismatch over the past few decades with natural variation alone (Case a) is easily discernible. This and several related considerations led to the conclusion: "In light of new evidence and taking into account the remaining uncertainties, most of the observed warming over the past 50 years is likely to have been due to the increase in greenhouse gas concentrations” (p. 10, IPCC Summary For Policymakers). The IPCC researchers used the word "likely" to represent a judgmental estimate of confidence level of a 66-90% chance of being correct (p. 2, Footnote 7, IPCC Summary For Policymakers).

Question: (Directed to Dr. Albritton) If a scientist wanted to "prove" that the warming we've seen is, in fact, just part of the background noise and not caused by people, what kind of proof would he need? And is anyone looking for that kind of proof? Answer:

In the approach outlined in the last question regarding natural fluctuations, a study would have to demonstrate that an hypothesized natural process, when added to a simulation of surface temperature, matched the observed temperature record, to a specified confidence level. Several such natural mechanisms (e.g., changes in total solar irradiance, solar ultraviolet radiation, cosmic rays and clouds, and volcanic emissions) have been hypothesized. As the IPCC researchers assessed, they have not produced reliable simulations of the warming of the past 50 years. It was noted that solar irradiance changes may have contributed to the observed warming in the first half of the 20th century. Sulfate particles from the emissions of explosive

volcanoes (e.g., Mt. Pinatubo in 1991) have been observed to cause a small cooling of the climate system for a few years until the particles have settled out of the atmosphere. No such test has found a natural process that could simulate the warming of the past 50 years.

Hypotheses of potential new climate-relevant processes, both natural and humaninfluenced, will no doubt continue to be put forward and tested, since that is the scientific process by which understanding is improved.

Question: (Directed to Dr. Albritton) Would you say that the IPCC's findings "prove" that climate change is being caused by human activities? If not, what scientific "proof" would be required, and what do you think is necessary to get that proof? Answer:

Scientific insights are described as a conclusion with a stated confidence level. As noted under Question (A) above, the researchers of the detection/attribution chapter (Chapter 12) of the IPCC placed a 66-90% confidence level in the attribution that most of the warming observed over the past 50 years is due to human activities. As the three major assessments by IPCC researchers over the past decade indicate, the confidence level associated with the detection of a climate change and the attribution to human influences has increased over the last 10 years. The reasons are several fold: a longer and more closely scrutinized temperature record, better simulation of natural climate variations, and new estimates of the climate response to natural and human-influence climate forcings. Further, the magnitude of the greenhouse-gas forcing of climate-altering radiation increases each year; therefore, the ease of detection increases with time.

For reference, the earlier and current attribution conclusions of IPCC researchers

are:

IPCC (1990): "The size of this warming [0.3-0.6 degrees Celsius over the last 100 years] is broadly consistent with predictions of climate models, but it is also of the same magnitude as natural climate variability. ... The unequivocal detection of the enhanced greenhouse effect is not likely for a decade or more." IPCC (1995): "The balance of evidence suggests a discernible human influence on global climate."

IPCC (2001): "In light of new evidence and taking into account the remaining uncertainties, most of the observed warming over the past 50 years is likely [i.e., 66– 90% confidence level] to have been due to the increase in greenhouse gas concentrations."

How can the understanding of climate changes and their causes be improved, that is, a statement with yet-higher confidence? The IPCC researchers identified the source of remaining uncertainties in detection and attribution. As noted, many of the explicit near-term research needs summarized above address improvements in detection and attribution (e.g., discrepancies between the vertical profile of temperature change in the lower atmosphere seen in observations and simulated by climate models).

PREPARED STATEMENT OF BERRIEN MOORE III

I. INTRODUCTION

There has been encouraging progress over the past decade. We understand better the coupling of the atmosphere and ocean. Significant steps have been taken in linking the atmosphere and the terrestrial systems, though the focus tends to be on water-energy and the biosphere with fixed vegetation patterns. There is also encouraging progress in developing integrated-assessment models that couple economic activity, with associated emissions and impacts, with models of the biogeochemical and climate systems. This work has yielded preliminary insights into system behavior and key policy-relevant uncertainties.

The challenges are significant, but the record of progress suggests that within the next decade the scientific community will develop fully coupled dynamical (prognostic) models of the full Earth system (e.g., the coupled physical climate, biogeochemical, human subsystems) that can be employed on multi-decadal time scales and at spatial scales relevant to strategic impact assessment. Future models should certainly advance in completeness and sophistication; however, the key will be to demonstrate some degree of prognostic skill. The strategy will be to couple the biogeochemical-physical climate system to representations of key aspects of the human system, and then to develop more coherent scenarios of human actions in the context of feedbacks from the biogeochemical-physical climate system.

Developing these coupled models is an important step. From the perspective of understanding the Earth system, determining the nature of the link between the biogeochemical system and the physical-climate system represents a fundamental scientific goal. Present understanding is incomplete, and a successful attack will require extensive interdisciplinary collaboration. It will also require global data that clearly documents the state of the system and how that state is changing as well as observations to illuminate more clearly important processes.

II.1 Overview

II. THE CLIMATE SYSTEM

Models of physical processes in the ocean and atmosphere provide much of our current understanding of future climate change. They incorporate the contributions of atmospheric dynamics and thermodynamics through the methods of computational fluid dynamics. This approach was initially developed in the 1950s to provide an objective numerical approach to weather prediction. It is sometimes forgotten that the early development of "supercomputers" at that time was motivated in large part by the need to solve this problem. In the 1960s, versions of these weather prediction models were developed to study the "general circulation" of the atmosphere, i.e., the physical statistics of weather systems satisfying requirements of conservation of mass, momentum, and energy. To obtain realistic simulations, it was found necessary to include additional energy sources and sinks: in particular, energy exchanges with the surface and moist atmospheric processes with the attendant latent heat release and radiative heat inputs.

Development of models for the general circulation of the ocean started later, but has proceeded in a similar manner. Models that deal with the physics of the oceans have been developed and linked to models of the atmospheric system. Within ocean models, the inclusion of biogeochemical interactions has begun, with a focus upon the carbon cycle. Modelling of the biological system, however, has been more challenging, and it has been only of late that primitive ecosystem models have been incorporated in global general circulation ocean models. Even though progress has been significant, much remains to be done. Eddy-resolving ocean models with chemistry and biology need to be tested and validated in a transient mode, and the prognostic aspects of marine ecosystems including nutrient dynamics need greater attention at basin and global scales.

Model development for the ocean and atmosphere has had a fundamental theoretical advantage: It is based on the firmly established hydrodynamic equations. At present there is less theoretical basis for a "first principles" development of the dynamical behavior of the terrestrial system. There is a need to develop a fundamental methodology to describe this very heterogeneous and complex system. For the moment, it is necessary to rely heavily upon parameterisations and empirical relationships. Such reliance is data intensive and hence independent validation of terrestrial system models is problematical. In spite of these difficulties, a coordinated strategy has been developed to improve estimates of terrestrial primary productivity and respiration by means of measurement and modelling. The strategy has begun to yield dividends. Techniques from statistical mechanics have been wedded with biogeochemistry and population ecology yielding new vegetation dynamic models.