National Academies Press: OpenBook
« Previous: Part II Detailed Supporting Analysis
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 51
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 52
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 53
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 54
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 55
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 56
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 57
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 58
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 59
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 60
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 61
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 62
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 63
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 64
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 65
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 66
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 67
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 68
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 69
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 70
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 71
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 72
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 73
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 74
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 75
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 76
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 77
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 78
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 79
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 80
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 81
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 82
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 83
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 84
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 85
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 86
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 87
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 88
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 89
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 90
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 91
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 92
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 93
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 94
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 95
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 96
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 97
Suggested Citation:"4 Progress Toward the Research Elements." National Research Council. 2007. Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results. Washington, DC: The National Academies Press. doi: 10.17226/11934.
×
Page 98

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

4 Progress Toward the Research Elements T his chapter presents the committee’s stage 1 analysis of the Climate Change Science Program (CCSP) research elements. The preliminary assessment was structured around the matrix (Appendix C), which evaluates progress of the 33 research questions in five categories: (1) data and physical quantities, (2) understanding and representation of processes, (3) uncertainty, predictability, or predictive capabilities, (4) synthesis and assessment, and (5) risk management and decision support. The goal was to highlight the most important issues, as identified by the peer-review workshop, not to provide an exhaustive analysis of every aspect of each research question. Consequently, although scores and commentary were assigned to all 165 cells in the matrix, this chapter reports only an overall qualitative score (good, fair, inadequate) and key comments for each re- search question. The scores are defined as follows: • Good = The quality and contribution of work exceeds expectation. • Fair = The quality and contribution of work merely meets expecta- tion. Additional review may be warranted to increase effectiveness. • Inadequate = The quality and contribution of work does not meet the needs of the program. Additional review to explain the poor results is required. Recurring themes and trends are discussed under “Opportunities and Threats” for each research element. The chapter concludes with an example 51

52 EVALUATING PROGRESS OF THE U.S. CCSP of how progress in the overarching goals can be evaluated, based on scores for the relevant research questions. ATMOSPHERIC COMPOSITION The composition of the atmosphere plays a critical role in connect- ing human welfare with climate changes because the atmosphere links the principal components of the Earth system. Emissions of gases and particles from natural sources and human activities enter the atmosphere and are transported to other geographical locations and often to higher altitudes. Some emissions undergo chemical transformation or removal while in the atmosphere or influence cloud formation and precipitation. Changes in atmospheric composition alter the greenhouse effect and the reflection and absorption of solar radiation, which modifies the Earth’s radiative (energy) balance. Subsequent feedbacks and responses to this human-in- duced climate forcing influence human health and natural systems in a variety of ways. Observed trends in atmospheric composition are among the earliest harbingers of environmental change. Because the atmosphere acts as a long-term reservoir for certain trace gases, any associated global changes could persist for decades or even millennia, affecting all countries and populations. The CCSP approach to understanding the role of atmospheric composi- tion integrates long-term (multidecadal) systematic observations, laboratory and field studies, and modeling, with periodic assessments of understanding and significance to decision making. Most of the activities related to the atmospheric composition research element are carried out through national and international partnerships, partly because of the breadth and complex- ity of the science and policy issues and partly because the atmosphere links all nations. CCSP-supported research focuses on how the composition of the global atmosphere is altered by human activities and natural phenom- ena, and how such changes influence climate, ozone, ultraviolet (UV) radia- tion, pollutant exposure, ecosystems, and human health. Specific objectives address processes that affect the recovery of stratospheric ozone; properties and distributions of greenhouse gases and aerosols; long-range transport of pollutants and the implications for regional air quality; and integrated as- sessments of the effects of these changes. Interactions between atmospheric composition and climate variability and change, such as the potential effects of global climate change on regional air quality, are of particular interest. Progress Toward Answering the Research Questions In situ and satellite measurements and field campaigns have yielded rich data sets and improved estimates of physical quantities for all five questions

PROGRESS TOWARD THE RESEARCH ELEMENTS 53 under this research element. Gaps remain, however. Similarly, gains in our understanding and representation of many key physical processes have been substantial, although large uncertainties about the indirect effect of aerosols on climate, poor quantification of aerosol solar absorption, and the absence of aerosol-cloud-precipitation interactions in coupled models remain major shortcomings. Great uncertainties also remain in our knowledge of the radiative forcing of non-CO2 gases (e.g., tropospheric ozone). Although predictions of air quality and ozone have improved, the predictability of the impact of pollutants on human health and especially on ecosystems is still limited. Finally, although we have sufficient understanding of some atmospheric species (e.g., sulfates and nitrates) to promote action, the same is not true for other aerosols (e.g., elemental and organic carbon) and non- CO2 greenhouse gases. Q 3.1. What are the climate-relevant chemical, microphysical, and optical properties, and spatial and temporal distributions, of human-caused and naturally occurring aerosols? Good scientific progress has been made on several fronts (e.g., observa- tionally constrained aerosol direct forcing), but large uncertainties remain (e.g., emission sources, indirect effect of aerosols on climate). Progress in aerosol observations has enabled the Intergovernmental Panel on Climate Change (IPCC) to quantify for the first time the net contribution of aerosols to anthropogenic forcing (IPCC, 2007). Significant CCSP investments in this question reflect growing recognition of the importance of aerosols and their role in climate. A wealth of new data from space and ground measurements have been used effectively to generate physical properties such as aerosol absorption and anthropogenic fraction on a global scale. These data sets provided the first information on how aerosols are transported from land regions to oceanic regions. For example, the CCSP sponsored field experi- ments on transport and transformation processes in aerosol plumes off the east coasts of Asia and North America. Data from ground stations in the western United States have shown that springtime background aerosol in that region is Asian in origin (Heald et al., 2006b). The Indian Ocean Ex- periment and the Asian Pacific Regional Aerosol Characterization Experi- ment field campaign revealed that satellite-derived maps of aerosol optical depth and aerosol mixture (air-mass type) extent, combined with targeted in situ component microphysical property measurements, can provide a detailed global picture of aerosol properties and distributions and their direct radiative forcing (Chung et al., 2005; Yu et al., 2006). Such investiga- tions provided the first observationally constrained estimates of the effect of anthropogenic aerosols on climate. Another major advance is the first measurement of the effect of aerosols, including sunlight-absorbing black

54 EVALUATING PROGRESS OF THE U.S. CCSP carbon (soot), on the inhibition of cloud formation by the Moderate-Reso- lution Imaging Spectroradiometer (MODIS) (Kaufman et al., 2005a, b). There is still room for improvement, however. Large uncertainties remain about the emission sources of elemental and organic carbon, the indirect effect of aerosols on climate, and the extent of atmospheric solar absorption. Incorporation of aerosol-cloud interactions in coupled models has been slow. The CCSP has not undertaken a coordinated effort to evalu- ate future scenarios of changes in worldwide aerosol emissions, which seri- ously limits projections of future climate changes. Improved knowledge of aerosol forcing would have a major impact on decision support systems and policy actions: reductions in aerosols would reduce the aerosol masking ef- fect on global warming and accelerate greenhouse forcing over the next few decades. However, no coordinated efforts to provide information to climate modelers or to policy makers are apparent. Finally, understanding of some types of aerosols (e.g., sulfates, nitrates, elemental carbon) is insufficient to evaluate and promote specific actions. Q 3.2. What are the atmospheric sources and sinks of the greenhouse gases other than CO2 and the implications for the Earth’s energy balance? Good progress has been made on radiative forcing and sources and sinks of some greenhouse gases, such as methane, but uncertainties re- main for other greenhouse gases, limiting progress on decision support. Good measurements exist of most non-CO2 greenhouse gases (e.g., nitrous oxide, chlorofluorocarbons [CFCs], methane, carbon monoxide, ozone, hydrogen, hydrochlorofluorocarbons, hydrofluorocarbons, methyl halides, sulfur hexafluoride), although better measurements are required for some. Analyses have shown, for instance, that global methane abundances were constant for nearly seven years beginning in 1999, suggesting that methane may have reached a steady state in the atmosphere for reasons that are not yet known (Dlugokencky et al., 2003). The Aura satellite is providing the first-ever daily global measurements of tropospheric ozone and many other trace gases with unprecedented spatial resolution. A 350-year history of atmospheric carbonyl sulfide from an Antarctic ice core and firn air showed how atmospheric abundances of this gas have changed as a consequence of industrial sulfur emissions (Aydin et al., 2002). Researchers have also made good progress in understanding North American emissions of trace gases, which are precursors of the formation of aerosols and ozone (Heald et al., 2006a; Pfister et al., 2006). However, although good measurements exist, large uncertainties re- main in our knowledge of the radiative forcing of non-CO2 gases. Similarly, while the sources and sinks of many of these gases are better understood,

PROGRESS TOWARD THE RESEARCH ELEMENTS 55 unanswered questions on emission and removal processes remain. Many non-CO2 greenhouse gases are not yet included in global climate models. Q 3.3. What are the effects of regional pollution on the global atmosphere and the effects of global climate and chemical change on regional air quality and atmospheric chemical inputs to ecosystems? Good progress has been made in describing the fate of anthropogenic emissions in the global atmosphere through new measurement techniques and observational studies, yet predictability is still limited. Considerable work has been done on this question. For instance, the National Aeronau- tics and Space Administration’s (NASA’s) Transport and Chemical Evolu- tion over the Pacific mission demonstrated the value of global satellite and airborne observations for improving knowledge of emissions inventories (Streets et al., 2003; Wang et al., 2005). Broad-based initiatives to study anthropogenic emissions in megacities are now under way (Guttikunda et al., 2005; Madronich, 2006). Data from the Aura satellite are being used to help monitor pollution production and transport between cities, regions, and continents on a daily basis for the first time. The International Consor- tium for Atmospheric Research on Transport and Transformation carried out the largest climate and air quality study to date, with a focus on devel- oping a better understanding of the factors involved in the intercontinental transport of pollution and the radiation balance in North America and the North Atlantic (Singh et al., 2006). Finally, a new technique that enables measurement of trace gases in the atmosphere has opened a new frontier on the atmospheric chemistry that occurs at night. Nighttime reactions in- volving nitrogen-containing trace gases can effectively remove these gases from the atmosphere, and “short-circuit” the chemical reactions that would have produced ozone the next day (Sillman et al., 2002; Ren et al., 2003). Although the understanding of atmospheric chemistry processes and the impact of pollutants on human health has improved, a number of complexities, especially on the regional scale, limit predictability. Under- standing of the heterogeneous chemistry from local to global scales is still not sufficient to include in global models and make predictions of future changes. Finally, uncertainties remain about longer-term trends (e.g., for tropospheric ozone) that are important for interpreting the historical global climate record (Lamarque et al., 2005). Q 3.4. What are the characteristics of the recovery of the stratospheric ozone layer in response to declining abundances of ozone-depleting gases and increasing abundances of greenhouse gases?

56 EVALUATING PROGRESS OF THE U.S. CCSP The recovery of stratospheric ozone is a success story, where decisions were made despite some scientific uncertainty. Recent advances in under- standing and modeling stratospheric transport and dynamics have since reduced these uncertainties. The Scientific Assessment of Ozone Depletion (WMO, 2003) summarizes current understanding of the ozone layer and the phenomenon of stratospheric ozone depletion. CCSP-sponsored work continues to improve knowledge of the atmospheric processes underlying ozone abundance at the poles and globally, to support satellite observations of ozone-depleting substances in the atmosphere, to revise expectations for recovery of the ozone layer, and to develop approaches to evaluate the im- pacts of very short-lived halogen-containing substances on the ozone layer. Ground-based measurements of ozone are now sufficiently accurate to validate the satellite data, and their temporal resolution is sufficiently fine to determine diurnal variations and understand observed trends over the last century or more. For example, nine years of radiometer data from the UV-B Monitoring and Research Program’s observational network has been used to assess the geographic distribution, trends, and year-to-year variability of UV-B radiation in the United States (Grant and Slusser, 2004). Progress is inadequate, however, on the critical exchange processes between the troposphere and the stratosphere, the feedback mechanisms between increasing concentrations of greenhouse gases and reduced levels of chlorofluorocarbons, and predictions of the amount of water vapor in the stratosphere. Q 3.5. What are the couplings and feedback mechanisms among climate change, air pollution, and ozone layer depletion, and their relationship to the health of humans and ecosystems? Improved decadal and longer term climate and ozone data have driven good progress in the description of the effects of long-term changes in stratospheric and tropospheric temperatures and circulation on ozone-layer depletion. However, predictability is still limited because of insufficient un- derstanding of the couplings between air pollution and climate change. This broad and complex question is tailor-made for the CCSP because it is inher- ently interdisciplinary and requires strong interagency cooperation. Strong leadership has led managers of fragmented programs to pool resources in this arena. Resulting field programs engendered by the CCSP have yielded good results, and fair progress has been made in understanding processes that link long-term (several decades) changes in temperatures and circula- tion with ozone depletion. The impacts of pollutants on human health in New York City have been studied (Drewnick et al., 2004), yet significant gaps remain in understanding the connections between atmospheric compo-

PROGRESS TOWARD THE RESEARCH ELEMENTS 57 sition and human health, and especially between atmospheric composition and ecosystem health (NRC, 2001c). It is still not possible to model the full range of aerosol constituents in polluted areas, primarily because of the inherent complexity of the prob- lem and secondarily because concentrations of organic aerosols in urban environments are still uncertain by a factor of ten. Local- to global-scale heterogeneity of cloud processing of aerosols and the subsequent modifi- cation of aerosol chemistry also remain very uncertain. Work in this area has not reached the stage where scientific understanding can support risk management and decision making. Opportunities and Threats A large amount of high-quality satellite and in situ data, increasing computational resources, and sophisticated models have led to good prog- ress in understanding the factors that alter atmospheric composition and how these alterations affect climate, humans, and ecosystems. However, the absence of a well-coordinated national effort is limiting progress in improving aerosol emission strengths globally, estimating past histories of biomass burning, and determining the vertical distribution of aerosols and their solar absorption. Another major issue is that while satellite data are currently a rich resource, primarily because of the investment that began in the 1990s with NASA’s Earth Observing System, the future looks relatively bleak. The recent National Polar-orbiting Environmental Satellite System (NPOESS) downscale has eliminated several key climate instruments, such as the aero- sol polarization sensor (NRC, 2007a). Moreover, since most climate records require overlapping intercalibration to ensure accurate climate monitoring, future gaps in high-quality data will in many cases restart the climate record (NRC, 1998, 2004b; Trenberth et al., 2006). Such gaps are now likely, and since satellites require 5 to 10 years of advance planning, the NPOESS downscale must be dealt with soon. The future degradation of the climate data system is a problem for most of the CCSP science questions. CLIMATE VARIABILITY AND CHANGE Much has been learned over the past few decades about the Earth’s climate system components, the interactions among them and their vari- ability, and how and why the climate system is changing. This improved understanding is continually translated into better models of climate system components and of the fully coupled system, and these models are being applied to important scientific and societal questions. For example, current models indicate that the observed global-, continental-, and ocean basin-

58 EVALUATING PROGRESS OF THE U.S. CCSP scale temperature increases of the past several decades are outside the range of natural variability (IPCC, 2007). Observations underlie many of the advances in our understanding of the climate system. Ground-, ocean-, and space-based observations of key climate variables (e.g., surface and atmospheric temperature, precipitation, atmospheric moisture, clouds, winds, aerosols, sea level) provide insight on climate forcings (e.g., variations in solar output), processes (e.g., clouds, precipitation), and feedbacks (e.g., surface cover, albedo). Their compila- tion into long-term climate data records enables regional details of changes that are occurring in the global environment and their connections to hu- man activities to be discerned (Alverson and Baker, 2006; NRC, 2007a). Paleoclimate data sets enable assessment of longer-term variability within the climate system, and also place the global climate changes observed in recent decades within a longer context (NRC, 1990, 2006d). The climate variability and change research element plays an integra- tive role in the CCSP and is therefore central to the entire enterprise (CCSP, 2003). Specific objectives of the climate variability and change research element include reducing uncertainties and improving model predictions of climate variability and projections of change and determining their limits, assessing the likelihood of abrupt climate changes, examining how extreme events may be linked to climate variability and change, and formulating this knowledge in a way that can be integrated with non-climatic knowledge to support management and policy making. Progress Toward Answering the Research Questions Progress has been made in addressing the five questions under this research element, although accomplishments have been uneven. Better, longer, and more data sets have contributed to improved documentation and attribution of climate variability and change and to better understand- ing of many key climate processes (e.g., the global carbon cycle). However, as a result of ocean sampling limitations, evaluations of the decadal vari- ability in global heat content, salinity, and sea level changes can be made with only moderate confidence. Moreover, some processes (e.g., vertical ocean mixing, cloud feedbacks, the role of aerosols and ice sheet dynam- ics) are still relatively poorly understood. Although uncertainties remain in our understanding of climate processes and some processes need to be more fully represented (e.g., historical and likely future changes in land use; Feddema et al., 2005), state-of-the-art climate models are now able to reproduce many aspects of the climate of the past century, and simulations of the evolution of global surface temperature over the past millennium are consistent with paleoclimate reconstructions (IPCC, 2007). This achieve- ment improves confidence in future projections.

PROGRESS TOWARD THE RESEARCH ELEMENTS 59 Synthesis and assessment activities have also progressed (e.g., the release of Temperature Trends in the Lower Atmosphere; CCSP, 2006b), and some seasonal-to-interannual capabilities have been shared with stakeholders through the National Oceanic and Atmospheric Administration’s (NOAA’s) Regional Integrated Sciences and Assessments (RISA) program. However, although contributions to risk management and decision support have slowly increased, the individuals engaged have been few in number and many decisions have been made without strong scientific underpinnings. Q 4.1. To what extent can uncertainties in model projections due to climate system feedbacks be reduced? Investments in observation systems have paid off with improved under- standing and reduced uncertainties about feedbacks, although progress has been uneven and contributions to risk management and decision support have been inadequate. The response of global temperature to a given small forcing is proportional to the climate sensitivity. Feedback processes operat- ing in the atmosphere (e.g., changes in water vapor and cloud properties), ocean (e.g., efficiency of ocean mixing, changes in sea ice properties), and land (e.g., changes to surface cover, albedo, evapotranspiration, runoff, and biogeochemical cycles) collectively determine the climate sensitivity. The number and diversity of observations related to feedbacks have grown. For example, satellite data records over the past decade indicate that mass losses from the Greenland and West Antarctic ice sheets have contributed to global sea level rise (Velicogna and Wahr, 2006; Shepherd and Wingham, 2007) and that flow speed has been highly variable over short time inter- vals (a few years) for some Greenland outlet glaciers (Howat et al., 2007; Truffer and Fahnestock, 2007). Some key climate feedbacks (e.g., water vapor; see Trenberth, 2005) are better constrained, although less progress has been made on other im- portant feedbacks such as those involving ocean mixing (e.g., Wunsch and Ferrari, 2004), aerosol effects, and cloud processes. The initiation of climate process teams (CPTs) has encouraged much-needed collaboration between modelers and those involved in process- and observation-oriented research (see Chapter 5, “Modeling”), although CPT findings are just beginning to be incorporated into models (e.g., Danabasoglu et al., 2007). The availabil- ity of the suite of climate model simulations performed around the world to support the fourth IPCC assessment has resulted in a wider examination of climate system feedbacks, such as the sensitivity and response of polar systems to global climate change (e.g., Holland et al., 2006) and the pos- sible slowdown of the thermohaline circulation (Schmittner et al., 2005). However, scientific contributions to risk management and decision support have only begun to emerge.

60 EVALUATING PROGRESS OF THE U.S. CCSP Q 4.2. How can predictions of climate variability and projections of climate change be improved, and what are the limits of their predictability? Good progress has been made in improving the quality of climate model simulations of variability and change, although uncertainties remain, especially on local and regional scales, and inadequate progress has been made in using model predictions to support decision making. The best cli- mate models encapsulate the current understanding of physical processes involved in the climate system, their interactions, and the performance of the system as a whole. They have been extensively tested and evaluated using observations and have become useful instruments for carrying out numerical climate experiments. For example, climate model simulations that account for changes in both natural and anthropogenic climate forc- ings have reliably shown that the observed warming of recent decades is a response to increased concentrations of greenhouse gases and sulfate aerosols in the atmosphere (IPCC, 2007). Attribution studies have also demonstrated that many of the observed changes in indicators of climate extremes consistent with warming (e.g., annual number of frost days, warm and cold days, warm and cold nights) have likely occurred as a result of increased anthropogenic forcing (e.g., Tebaldi et al., 2005). Despite significant advances, climate models are not perfect, and some models are better than others. Uncertainties remain because of shortcom- ings in our understanding of climate processes operating in the atmosphere, ocean, land, and cryosphere and how to best represent those processes in models (e.g., Rodwell and Palmer, 2007). For example, parameterizations of vertical ocean mixing are unrealistic and most coupled ocean-atmosphere global circulation models mix heat into the ocean too efficiently (Forest et al., 2007). Moreover, the global coupled climate system exhibits a wide range of physical and dynamical phenomena with associated physical, bio- logical, and chemical feedbacks that collectively result in a continuum of temporal and spatial variability. The accuracy of predictions on time scales from days or seasons to years, as well as long-standing systematic errors in climate models, is limited by our inadequate understanding and capability to simulate the complex, multiscale interactions intrinsic to atmospheric and oceanic fluid motions (e.g., Meehl et al., 2001) and to represent all other unresolved small-scale processes in the ocean and at the land surface. For example, decadal climate predictions may require the initialization of coupled models with estimates of the observed state of the climate system. This initialization requires an ongoing commitment and strengthening of the observing system (Trenberth et al., 2002, 2006; GCOS, 2003; NRC, 2007a). However, although some observations and data networks have improved (e.g., Gravity Recovery and Climate Experiment [GRACE], Argo

PROGRESS TOWARD THE RESEARCH ELEMENTS 61 ocean profiling floats), others remain too sparse (e.g., atmospheric water vapor), poorly integrated with other essential observations (e.g., column ozone with temperature and water vapor), or in decline (e.g., Tropical Atmosphere Ocean [TAO] buoy array). Some observing systems suffer te- lemetry problems that have caused data to be lost (Trenberth et al., 2006). Ensembles of simulations that estimate the range of probable outcomes can be used to project climate change where uncertainty arises from limitations of the models and the emission scenarios used to represent the effects of human activity. Finally, use of climate model output by resource managers, planners, and decision makers remains limited, although exceptions exist. For ex- ample, a model of Lyme disease transmission, which simulates the effects of climate and other factors on disease risk, is being used by public health of- ficials to examine strategies for controlling tick populations (NRC, 2001c). However, the prediction value of such models is limited by uncertainties in the climate-disease relationship and the confounding influence of other factors. The California Department of Water Resources used a statistical analysis of VIC model outputs to obtain monthly average streamflows that could be used to estimate how reservoirs inflows would be affected by cli- mate change (CDWR, 2006). In general, however, resource managers need research results to be translated into different forms of information. Apart from programs such as RISA that have facilitated sharing of seasonal-to- interannual capabilities with stakeholders, few research results have been used to support risk management and decision making. Q 4.3. What is the likelihood of abrupt changes in the climate system such as the collapse of the ocean thermohaline circulation, inception of a de- cades-long mega-drought, or rapid melting of the major ice sheets? CCSP management has been effective in marshaling the necessary re- sources to help answer this question. Good progress has been made in docu- menting abrupt climate change, but predictive capability remains low and the impact on decision making has been minimal. Good progress has been made in documenting abrupt climate change (e.g., mega-droughts) from proxy records such as lake cores (Vershuren et al., 2000) ice cores (Thomp- son et al., 2000, 2006), and integrated tree ring and observational data (Herweijer et al., 2006). Longer and more comprehensive data sets have revealed evidence of past abrupt changes that have the potential to occur in the future (Trenberth et al., 2004; Kerr, 2005). A 300-year long drought similar to the one that gripped East Africa 4,000 years ago (Thompson   Since 1999 the number of Argo floats has increased to 2,856 out of about 3,000 planned. See <http://wo.jcommops.org/>.

62 EVALUATING PROGRESS OF THE U.S. CCSP et al., 2002) would have devastating consequences today. Of particular concern is that under warmer conditions, it is likely that heat waves and droughts will become both more severe and more frequent than those in the past (IPCC, 2007). Understanding and representing processes has improved significantly in some areas, but less so in others. For example, fair progress has been made in understanding mechanisms that force sustained drought (Trenberth et al., 2004; Kerr, 2005), but the likelihood of mega-droughts is unknown, despite the existence of efforts such as North American drought reconstructions. Both a collapse in the thermohaline circulation (Latif et al., 2000; Gregory et al., 2005) and a catastrophic release of methane hydrates (Schaefer et al., 2006) now seem unlikely. However, much more work (e.g., radar mapping of East Antarctica, dynamical modeling of the large ice sheets and their outlet glaciers) is needed to anticipate potentially large and abrupt changes in the climate system. The mechanisms of past abrupt climate changes are not yet fully un- derstood, and climate models typically underestimate the size, speed, and extent of those changes. Some processes (e.g., ice sheet sliding) represent major uncertainties in future climate projections (Vaughan and Arthern, 2007, and references therein). Abrupt changes are not predictable, although their past occurrence can be used to derive probabilistic estimates of future occurrence. However, such estimates depend on the assumption that the past was statistically similar to the present, which is unlikely given the unique climate we are currently experiencing. Given the state of knowledge, surprises are inevitable (Alley et al., 2003), and because of greenhouse warming and other human alterations of the Earth system, certain thresh- olds are likely to be crossed (IPCC, 2007). Despite wide media coverage, risks have not been quantified and the costs of undesirable surprises have not been factored into economic models (NRC, 2002a). Q 4.4. How are extreme events, such as droughts, floods, wildfires, heat waves, and hurricanes, related to climate variability and change? Fair progress has been made on producing (1) the longer histories, span- ning centuries to a few millennia or more, needed to advance understanding and to increase prediction of extreme events and (2) risk assessments. For any change in mean temperature, there is likely to be an amplified change in extremes (Tebaldi et al., 2005). Extreme events, such as heat waves and droughts, are exceedingly important to both natural and human systems. Humans are adapted to a range of weather conditions, but extremes of weather and climate exceed these tolerances. Widespread changes in tem- perature extremes have been observed over the last 50 years (Easterling et al., 2000; Alexander et al., 2006) and are expected to increase in frequency

PROGRESS TOWARD THE RESEARCH ELEMENTS 63 if the Earth continues to warm (IPCC, 2007, Table SPM2). In particular, the number of heat waves has increased globally, and increases in the number of warm nights have been widespread (Alexander et al., 2006). Cold days, cold nights, and days with frost have become rarer (e.g., Tebaldi et al., 2005, and references therein). Freezing levels have risen in elevation (Diaz et al., 2003), and the spring maximum snowpack is expected to diminish as the climate warms (Snover et al., 2003), reducing the spring runoff that supplies much of the streamflow of the western United States. Drying has been pronounced throughout the subtropics of both hemispheres (e.g., Ho- erling and Kumar, 2003; Seager et al., 2005), and the risk of more frequent and severe droughts has likely increased (Trenberth et al., 2004). Moreover, with global warming, warmer sea surface temperatures, sea level rise, and increased atmospheric moisture content mean that hurricanes will likely become more intense (e.g., Emanuel, 2005; Webster et al., 2005; Anthes et al., 2006). Modeling of many of these aspects has improved, but simulating small-scale extreme events remains a challenge. Climate extremes of the past are recorded in high-resolution histories extracted from ice cores, tree rings, and other kinds of paleoclimate re- cords. High-resolution paleotemperature histories from ice cores collected in Greenland (Masson-Delmotte et al., 2005; Mosley-Thompson et al., 2006) and Antarctica (Masson et al., 2000), as well as from tropical glaciers (Thompson et al., 2000, 2006) and extensive tree ring networks (Briffa et al., 2004; Osborn and Briffa, 2006) have revealed the differing tempera- ture trends and variability among geographic regions. However, similarly high-resolution records that sample a wider range of regions and record other extreme events (e.g., severe storms, monsoon failure) are lacking. The reduction in the number of surface observing stations in recent decades (GCOS, 2003) will negatively affect documentation of extreme events. The delay of the Global Precipitation Measurement (GPM) mission will likely lead to a gap in critical regional records used to predict high-impact weather events such as floods, droughts, and landslides (NRC, 2007a). CCSP synthesis and assessment product 3.3 (weather and climate ex- tremes) is now in draft form. A National Research Council review found that the draft report provides a thorough assessment of the key issues, although the discussion of drought and ecological impacts should be strengthened (NRC, 2007b). In addition to this focused effort, the insur- ance industry has gained a better awareness of the links between climate change and hurricanes (e.g., see Schiermeier, 2006). Risk assessments for some extreme events (e.g., drought, heat waves) are under way and heat wave preparedness has improved (Goodrich and Ellis, 2006). Knowledge of wildfires associated with climate variability and change is also beginning to inform forest management (Morehouse et al., 2006).

64 EVALUATING PROGRESS OF THE U.S. CCSP Q 4.5. How can information on climate variability and change be most efficiently developed, integrated with non-climatic knowledge, and com- municated in order to best serve societal needs? This question was difficult to assess, but it is clear that progress has been inadequate. A different formulation (e.g., How can information on climate variability and change best be communicated to intermediaries who provide tailored information to the public?) might enhance its evaluation. Opportunities and Threats Rich data collections are currently available for use in understanding processes, reducing uncertainties, and constraining models. These data form the backbone upon which continued advances will be made. Studies of climate variability and change require long, continuous, high-fidelity re- cords of key system variables. Consequently, maintenance of a suite of key observing systems is important. Unfortunately, many observing systems are put in place without a viable plan for follow-on support. In addition, new observing systems are required to meet new scientific needs. However, the number of satellite sensors is expected to decrease dramatically by the end of the decade (NRC, 2007a). The International Polar Year offers numerous opportunities for multinational collaborations (NRC, 2004d; Pennisi et al., 2007), but sustained funding for U.S. involvement is not assured (Leshner, 2007). As our knowledge of the different components of the climate system and their interactions has increased, so has the complexity of climate mod- els. Many of the most pressing scientific questions regarding the climate system and its response to natural and anthropogenic forcings cannot readily be addressed with traditional models of the physical climate. A key near-term climate change issue, for example, is the response of terrestrial ecosystems to increased concentrations of carbon dioxide. Will plants begin releasing carbon dioxide to the atmosphere in a warmer climate, thereby acting as a positive feedback, or will vegetation absorb more carbon dioxide and hence decelerate global warming? Related issues include the interac- tions among land use change, deforestation by biomass burning, emission of greenhouse gases and aerosols, weathering of rocks, carbon in soils, and marine biogeochemistry. Exploration of these questions requires a more comprehensive treat- ment of the integrative Earth system. In order to address these emerging issues, physical models are being extended to include the interactions of climate with biogeochemistry, atmospheric chemistry, ecosystems, glaciers and ice sheets, and anthropogenic environmental change. These new Earth system models, however, will require large investments in computing infra-

PROGRESS TOWARD THE RESEARCH ELEMENTS 65 structure before they can be fully utilized. Similarly, inadequate resources for computing are limiting progress in several key modeling areas, including representation of extremes and feedbacks and paleoclimate simulations, which are critical for testing Earth system behavior on climate-relevant time scales (see Chapter 5, “Modeling”). Finally, it is imperative that data archiving keep pace with data acqui- sition and advances in climate system understanding. Proper archiving of data to ensure their availability requires strong interagency cooperation, and the CCSP offers an excellent structure for such agreements and plans to be formulated. WATER CYCLE The water cycle encompasses the dynamics of water stored in the at- mosphere (clouds and water vapor), the ground (soil moisture and ground- water), and at the surface (snow and ice, lakes, and oceans), as well as the fluxes (e.g., precipitation, evapotranspiration, runoff, recharge) between these stores. The water cycle plays a direct role in agriculture, ecosystems, industry, and transportation, and it is inherently linked to the climate sys- tem through its interaction with the energy cycle. Climatically significant amounts of energy are stored and transported as sensible heat in the ocean and latent heat in the atmosphere, and the distribution of water in the at- mosphere and on the ground regulates radiative transmission and reflection and influences the energy balance of the land surface and boundary layer. The CCSP water cycle research element focuses on (1) quantification of the water cycle through observed and modeled budgets at local to global scales; (2) process-based understanding of the physics, chemistry, and biol- ogy involved in the water cycle and their interaction with other parts of the climate system; and (3) development of economically and socially relevant predictive capabilities of seasonal-to-interannual anomalies and geographi- cal shifts in the climate mean. The wide range of scales over which hydro- logic processes act and can be measured and modeled, as well as the wide range of basic sciences involved require coordinated efforts from agencies with different research expertise and interests as well as different observa- tional capabilities. Progress Toward Answering the Research Questions Progress in the water cycle research element has been uneven. Good progress has been made in understanding processes and improving models, built in part on data from satellites and field campaigns and strong leader- ship across participating agencies. Progress in bringing process understand- ing to bear on societal needs through improved predictive capabilities at

66 EVALUATING PROGRESS OF THE U.S. CCSP longer time scales and smaller spatial (e.g., regional) scales has been mixed. Intertwined with these success stories are examples of missed opportunities to communicate with stakeholders (e.g., through a CCSP synthesis and as- sessment product) or to stave off deterioration of data collection networks (e.g., streamflow, snowpack) required to improve predictions. Water cycle research and data collection are spread across many agencies, and CCSP leadership is essential for creating the types of coordinated interagency field campaigns and joint funding opportunities (e.g., Global Energy and Water Cycle Experiment Continental-scale International Project, World Ocean Circulation Experiment) that so successfully united scientists around research priorities in the 1980s and 1990s. Q 5.1. What are the mechanisms and processes responsible for the mainte- nance and variability of the water cycle; are the characteristics of the cycle changing and, if so, to what extent are human activities responsible for those changes? Progress has been mixed, with good advances in data collection, fair progress on predictability studies, and inadequate progress in understand- ing the impact of managed systems on the water cycle. Data from the Tropi- cal Rainfall Measuring Mission (TRMM) have greatly improved global precipitation mapping, and multisensor and combined in situ and satellite precipitation data products (e.g., from the Global Precipitation Climatology Project and GRACE) have begun to be developed. The use of data from the U.S. Department of Agriculture (USDA) Snowpack Telemetry (SNOTEL) network has improved snowpack-snow water equivalent maps and led to a better understanding of cold season and mountainous region hydrologic process. Estimates of natural flow and unimpaired river flow are important, but some stations in the U.S. Geological Survey (USGS) stream gauging network have been shut down. Some of these station records spanned mul- tiple decades and were valuable for understanding and assessing land use impacts on climate and water resources trends. Understanding of the mechanisms that control water fluxes among the components of the Earth system has improved over the last several years. Such understanding provides a critical link between anthropogenic climate forcing and impacts on human systems. However, current Earth system models do not include the contribution of managed systems such as agri- culture. The use of Earth system models for synthesis and assessment will also require inclusion of two-way coupling between such systems and the traditional components of climate models. Finally, a number of studies on predictability and limits of prediction, use of ensemble predictions (e.g., Hydrologic Ensemble Prediction Experiment), and analyses of uncertainty

PROGRESS TOWARD THE RESEARCH ELEMENTS 67 propagation have been initiated, but they are too new for progress to be evaluated. Q 5.2. How do feedback processes control the interactions between the global water cycle and other parts of the climate system (e.g., carbon cycle, energy), and how are these feedbacks changing over time? Progress toward answering this question has been fair. Satellite obser- vations have made positive contributions to the consistent and continuous data record needed to understand how feedbacks work between different components of the Earth system. However, plans for replacing observing systems such as the A-Train and TRMM are in doubt. Progress on data assimilation (merging models and observations) has been good, especially on the atmosphere (e.g., National Centers for Environmental Prediction- National Center for Atmospheric Research reanalysis) and ocean (e.g., Estimating the Circulation and Climate of the Ocean, Global Ocean Ecosys- tems Dynamics) components of this question. However, data assimilation efforts do not yet include geochemistry, especially of carbon. Good progress has also been made in understanding process such as cloud formation, air-sea interaction, and land-atmosphere interaction. However, additional work needs to be done on feedbacks that cut across physical processes, such as aerosol and moist processes and water and carbon fluxes. The strong leadership of program managers for multidisci- plinary and large community programs has been a key factor in many of these gains, and this leadership must be maintained or progress will stall. Finally, few efforts have been made to synthesize and assess water cycle research results for the broader stakeholder community. Q 5.3. What are the key uncertainties in seasonal-to-interannual predictions and long-term projections of water cycle variables, and what improvements are needed in global and regional models to reduce these uncertainties? Progress in predictability has been fair on seasonal-to-interannual time scales and inadequate on decadal and longer time scales. Less progress has been made on understanding long-term change and decadal variability than on interseasonal-to-interannual predictability because of the deficiency of sustained observing systems and robust models. The TAO array was started during the decade of the Tropical Ocean Global Atmosphere program (1985-1994) and has since been enhanced by the CCSP. These sustained observations have helped to produce modest advances in our understanding and ability to predict El Niño-Southern Oscillation (ENSO) in the tropical Pacific (NRC, 1996; Kondrashov et al., 2005). However, similar in situ

68 EVALUATING PROGRESS OF THE U.S. CCSP networks have only begun to be deployed in the Atlantic or Indian Oceans, and seasonal-to-interannual predictability is poor in those regions. Where ENSO signals exist, the seasonal water-related influence is well known. However, understanding remains poor in the inner continents. Be- cause soil moisture is difficult to measure at regional scales, most inferred predictability is based on model experiments (e.g., Koster et al., 2004). Although these are valuable, they are an inherently poor substitute for data- based validations of predictions. NASA’s Hydrosphere State (Hydros) mis- sion has been cancelled. Snowpack information is being collected through the USDA SNOTEL network, but snowpack water equivalent remote sens- ing mapping remains a deficiency. The GRACE mission has yielded the first depth-integrated water storage measurements (Swenson et al., 2006). Such estimates have already provided valuable checks on climate and weather models (Hirschi et al., 2006; Niu and Yang, 2006). Numerous well-wa- ter heights have been measured by local, state, and federal agencies for groundwater management and water rights administration. Knowledge of groundwater fluctuations over large scales is critical to improving represen- tation of groundwater processes in climate models. Groundwater storage and release to the surface are long-memory processes with potential to improve forecasting (Bierkens and van den Hurk, 2007). More systematic collection, standardization, and archiving of these data would help these advances to continue. Contributions to interannual predictability, such as the Pacific Decadal Oscillation and the North Atlantic Oscillation, have received increased attention and the physical processes are becoming better understood. However, models to predict their future behavior have not yet been fully evaluated. Probabilistic seasonal climate forecasts can be used to make probabilistic forecasts of malaria incidence (Thomson et al., 2005, 2006), and seasonal models have been used in applications ranging from agricul- ture to public health (e.g., Solow et al., 1998; Changnon, 2004). Q 5.4. What are the consequences over a range of space and time scales of water cycle variability and change for human societies and ecosystems, and how do they interact with the Earth system to affect sediment transport and nutrient and biogeochemical cycles? Progress toward answering these questions has been inadequate. Fun- damental data on water quality and sediment fluxes are not available outside of several experimental sites (e.g., Long-Term Ecological Research [LTER] sites). Other sources of information are geographically spotty (e.g., National Water-Quality Assessment Program), are useful for process studies but not monitoring, or have other limitations. For example, floodplain and coastal maps are not sufficiently precise to be used in process and assess-

PROGRESS TOWARD THE RESEARCH ELEMENTS 69 ment models. As a result, great uncertainty remains in understanding the dynamics of nutrient and sediment outflow (e.g., episodic or continuous). An increasing number of interdisciplinary research initiatives, such as in- tegrated water and carbon cycle research, have been proposed, but diverse specialists are not yet working together routinely. Q 5.5. How can global water cycle information be used to inform deci- sion processes in the context of changing water resource conditions and policies? Despite a few successes, overall progress has been inadequate in all aspects of the program, including data, predictive ability, and communica- tion. A number of critical water cycle measurements (e.g., precipitation, soil moisture, small-scale sea level topography) are needed to construct and test models that are used to make projections on water resources. However, three planned measurement systems (GPM, Hydros, Wide-Swath Altimetry) have been delayed or eliminated during the past few years. Accurate pre- dictions of sea level rise, based on measurements from satellites and tide gauges, are especially important because of the impacts on coastal urban areas, wetlands, and sensitive ecosystems. Uncertainties about the timing and magnitude of sea level rise are being narrowed, and scientists now have greater confidence in projections (IPCC, 2007). Some water resource man- agers are using outputs from regional models, but they need more refined models (e.g., at the watershed level) with improved resolution. The CCSP does not maintain an inventory of resources that might be useful to water resource managers and policy makers, nor does it com- municate with these stakeholders through newsletters or liaisons. A few programs (e.g., RISA, Columbia International Research Institute for Cli- mate and Society) have made fair progress in communicating the impacts and uncertainties of interannual variability and climate change on water resources. However, it seems likely that most water authorities and manage- ment offices have little knowledge of what the CCSP can offer. Opportunities and Threats Future progress in the water cycle research element would be fostered by stronger leadership and interagency coordination. For example, future sites of USDA’s Soil Climate Analysis Network program, which measures soil moisture and soil temperature profiles, could be collocated with the Department of Energy’s (DOE’s) AmeriFlux stations, which measure carbon and water fluxes in the boundary layer. Measurements of these variables at the same locations would improve our understanding of processes and increase our ability to model mass and energy exchanges between the land

70 EVALUATING PROGRESS OF THE U.S. CCSP surface and the atmosphere. Modeling of these exchanges has long been a weak and poorly constrained element in global climate models, due in large part to a scarcity of long-term data sets. This kind of budget-neutral coordination activity could increase the value of government investments in observing systems. The cancellation of critical observing systems has the potential to greatly slow future progress. For example, the Hydros soil moisture mission, which is important for answering all five water cycle research questions, was re- cently cancelled. The GPM mission is now delayed, jeopardizing continued progress in global precipitation mapping. It is unclear whether international (e.g., Global Earth Observing System of Systems) or foreign initiatives can take the place of cancelled missions. At the same time, surface-based data collection systems are either deteriorating (e.g., USGS streamflow monitor- ing) or continually threatened with cutbacks (e.g., USDA SNOTEL obser- vation system). Finally, the CCSP has no synthesis and assessment product to focus efforts on the water cycle, although product 5.3 (seasonal to interannual forecasts) is aimed primarily at water resource managers. Such efforts are especially important with the retirement of program managers who provided strong leadership in multidisciplinary and large-community pro- grams. A synthesis and assessment product could also provide a vehicle for communicating research results to water resource managers and policy makers. LAND USE AND LAND COVER CHANGE The land use and land cover change research element was created under the CCSP in 2002. Seventy percent of the funding for this research element is currently provided by NASA (CCSP, 2006b) and this is reflected in the research emphasis to date. NASA is the only agency on the Interagency Working Group (IWG) for Land Use and Land Cover Change with a formal land cover program. Program managers from other agencies on the IWG have little authority to commit resources to the research questions or to steer research directions. Land use and land cover changes are the most proximate and visible forms of global environmental change. Land use change occurs locally but is also significant at regional and global scales. For example, extensive tropical deforestation affects the global carbon cycle and thus the climate. At regional and local scales, deforestation can affect water quality, biodi- versity, and human livelihood. A variable and changing climate influences the distribution of land cover and the sustainability of land use practices, which in turn affect agricultural productivity, food supply, and human vulnerability in marginal lands. Urbanization and suburban extensification

PROGRESS TOWARD THE RESEARCH ELEMENTS 71 magnify these problems and create new ones for air and water quality, human health, and transportation systems. Fire at the wildland-suburban interface is a serious hazard to some ecosystems and a human health prob- lem in areas with extensive prescribed biomass burning. The impact of land use change on the resilience of human and social systems to climate change can be positive or negative. Land use manage- ment provides a means by which to adapt to the impacts of climate change (Pyke and Andelman, 2007). Consequently, predictions of land use and land cover change under a combination of climatic and economic scenarios are important for land use planning and policy at local, regional, and national scales. A close integration of natural and social science is required to ad- vance this research element. Progress Toward Answering the Research Questions A strong intellectual and technological foundation has been developed for this research element. Good progress has been made in the quantification and characterization of land use and land cover change, based in large part on the use of satellite data and improvements in the analysis of geospatial data (Walsh and Crews-Meyer, 2002). Social and biophysical processes are beginning to be combined in models, and considerable potential for growth exists for modeling climate and land use interactions. Progress in develop- ing synthesis and assessments and decision support has been inadequate, although a few research results have been synthesized, and some research has directly supported decision making (Rindfuss et al., 2004a). Q 6.1. What tools or methods are needed to better characterize historic and current land use and land cover attributes and dynamics? Good progress has been made on the characterization of land cover and its attributes and dynamics. The current need is to transition the methods developed in the research domain into the operational arena to obtain rou- tine and consistent global monitoring of land cover change at high (30 m) spatial resolution. The availability of satellite data has given the program considerable momentum and has fostered good progress in the study of land cover change, global mapping of land cover types, and regional mapping of local changes. Time-series analysis approaches developed in the 1990s have been refined, and improved land cover products have been generated from MODIS to study changes in global phenology, length of growing season, fire distributions, and tree cover (Friedl et al., 2002; Hansen et al., 2002; Zhang et al., 2004; Giglio et al., 2006). These new moderate-resolution (250 m) satellite data provided the means to detect land cover changes around the world (Zhan et al., 2002). Techniques for mapping large areas

72 EVALUATING PROGRESS OF THE U.S. CCSP of the Amazon using high-resolution (30 m) satellite data have been ex- tended to other tropical regions, improving estimates of the rates of land cover change (Skole and Tucker, 1993; Curran et al., 2004). The Landsat series of satellites has populated the national archive with unprecedented volumes of high-resolution data that facilitate continental- and regional-scale studies of land use (Arvidson et al., 2006). A national study of land cover change based on these data is currently under way. Early results for the eastern U.S. show that land cover change is associated primarily with an increase in timber harvesting and urban growth and a decline in agricultural activity. Methods have been developed to map the extent and changes in impervious surfaces and to model future urban devel- opment (Jantz et al., 2003). Methods have also been developed to include pre-satellite land use change using a combination of census population, housing, and agricultural data (e.g., Aspinall, 2004; Brown et al., 2005). At both global and regional scales, efforts have been made to compile his- torical data on agricultural land use extent (e.g., Ramankutty and Foley, 1999). However, in general less emphasis has been given to the historical record than to recent changes. Rapid land cover changes driven by major economic changes have been documented in a number of countries, including China and Paraguay (Seto and Kaufmann, 2003; Huang et al., 2007). Rates, causes, and conse- quences of urban land use change in the United States have been compiled by Acevedo et al. (2006). Studies are also assessing the nature, extent, and impact of land use change around areas of particular importance, such as conservation-protected areas, through, for example, shifting agriculture and selective logging (e.g., Curran et al., 2004; DeFries et al., 2005). Given the impact of land use change on biodiversity loss, particularly in the tropics, studying the causes, trends, and projected land use change around protected areas will become increasingly important. Q 6.2. What are the primary drivers of land-use and land-cover change? Fair progress has been made toward understanding the causes and process of land use and land cover change. Future progress will require greater emphasis on understanding the process of change in different physi- cal and social environments, and the development of general rules that can be used in land use modeling studies. A few studies have provided insights on the causes of land use change in a limited number of environments (e.g., Geoghegan et al., 2001; Fox et al., 2003). Progress has been hampered in part by the difficulty of obtaining socioeconomic data at the local scale and site-specific census data. Mechanisms must be found that enable provision   See <http://eros.usgs.gov/LT/LCCEUS.html>.

PROGRESS TOWARD THE RESEARCH ELEMENTS 73 and analyses of the data without compromising the privacy of individuals. A number of synthetic reviews have been carried out to develop general rules from a series of case studies (Rindfuss et al., 2004b; Hansen and Brown, 2005). Finally, the Large-Scale Biosphere-Atmosphere Experiment in Amazonia project has provided an in-depth look at selected land use pro- cesses (e.g., land use trajectories, logging practices, pasture development) in the Amazon and their role in the carbon cycle (Bierregaard et al., 2001; Roberts et al., 2003; Asner et al., 2005). Q 6.3. What will land use and land cover patterns and characteristics be 5 to 50 years into the future? Fair progress is being made toward future projections of land use change, but methods for modeling land use change could benefit from the development of community standards and best practices. Improved understanding of the processes of change is enabling predictive modeling of land cover changes (e.g., Moran and Ostrom, 2005). It is now time to start integrating interactive land use change processes and models with dynamic vegetation and climate models at the regional and global scales to examine the feedbacks. Studies in tropical regions have combined satel- lite data with household surveys to project future changes at the regional scale (e.g., Geoghegan et al., 2001). Regional-scale projections of land use change provide scenarios of future changes intended to be useful for in- forming policy (Zhang et al., 2006). In this context, a model of future land cover changes in the Amazon developed by Laurance et al. (2001) initiated considerable debate on the causes of deforestation and on modeling ap- proaches. Comparisons of the different modeling approaches has revealed methodological issues that have to be refined (Irwin and Geoghegan, 2001; Claggett et al., 2004). Land use modeling is in its infancy but is gathering momentum and a number of approaches and techniques are available or under development (Verburg and Veldkamp, 2005). In general, model applications have lacked rigor, and more attention has to be paid to uncertainty in prediction and to model validation. Dynamic, process-driven models of land use have yet to be integrated in ecosystem and climate modeling studies and feedbacks between the coupled systems have not been well investigated. Finally, no standard modeling tools exist for stakeholders interested in developing pro- jections of future land use change in the context of a changing climate. Q 6.4. How do climate variability and change affect land use and land cover, and what are the potential feedbacks of changes in land use and land cover to climate?

74 EVALUATING PROGRESS OF THE U.S. CCSP Inadequate progress has been made on this question. The role of land use in the carbon cycle and its contribution to greenhouse gas emissions has been understood for some time (IPCC, 2000). The effect of climate variability on land use is an area of continuing research and development, particularly in the context of ENSO predictions (e.g., Reilly et al., 2003). Researchers are only now beginning to quantify the impacts of land use change on climate and to understand the impacts of regional climate change on land use (e.g., DeFries et al., 2002; Marshall et al., 2004; Pielke, 2005). Feedbacks between land use change and climate are being studied at the regional scale (e.g., in the Amazon; see Laurance and Williamson, 2001). Relatively little funding has been devoted to this research area to date in the context of adaptation, and there is considerable potential for growth. Questions on how land use practices can be used to mediate the impacts of climate change and to sustain human livelihoods warrant further investiga- tion, not only in terms of mitigation but also of human vulnerability. Q 6.5. What are the environmental, social, economic, and human health consequences of current and potential land use and land cover change over the next 5 to 50 years? Although fair progress has been made toward answering this question, a wider range of impact studies focusing on human health and vulnerability is needed. Addressing this question requires significant interaction among CCSP research elements. Linkages between land use management and the environment are currently being explored by the CCSP. For example, land use in the context of the carbon and water cycles is part of the North American Carbon Program (NACP) (see “Carbon Cycle” below). A number of studies have examined the impacts of recent land use change, and the broader role of land use in the global carbon cycle continues to be refined (Houghton, 2003). Study of the impact of tropical land use on the carbon cycle and biodiversity continues to be hampered by the absence of continu- ous and systematic monitoring of land cover change in these ecosystems (DeFries et al., 2006). The global impact of land use change on the provi- sion of ecosystem services is starting to be recognized (Foley et al., 2005). The impacts of land use practices on water quality are being examined in different ecosystems (e.g., Mustard and Fisher, 2004). The effects of different aspects of land use on human health is being investigated (e.g., Patz et al., 2004; Kelly-Schwartz et al., 2004), but to date these studies have been supported almost entirely outside of the CCSP and have not benefited from interaction with the climate change community. The recent inclusion of the National Institutes of Health on the IWG for Land Use and Land Cover Change should help this integration. The paucity of CCSP research on impacts of climate change in general (see “Human

PROGRESS TOWARD THE RESEARCH ELEMENTS 75 Contributions and Responses to Environmental Change” below) has pre- vented significant progress on the societal impacts of land use change and human vulnerability. Opportunities and Threats Future progress in the land use and land cover research element is likely to be slow because of relatively low funding levels and the inability of the program to date to direct resources to strengthen the social science aspects of the research (e.g., process studies, research on societal adaptation). In- creased support to collect and integrate socioeconomic and environmental data could advance understanding of the consequences of land use change on human health (NRC, 2006a). Similarly, research on the effect of dif- ferent land use practices on the carbon budget (e.g., averted deforesta- tion, agro-forestry, no-tillage practices) could inform carbon management strategies. The groundwork has been laid for a community land use modeling initiative aimed at improving the representation of dynamic land use pro- cesses in climate and ecosystem models. The first step in developing such an initiative would be a review of current approaches to land use modeling and identification of best practices. The integration of ground-based mea- surements and high-temporal-frequency Landsat-class observations into regional-scale agriculture models would improve the accuracy of predic- tions (IGOL, 2006). The results of research on land use and land cover change are being used to inform land management policy, but more could be done with a close and sustained working relationship with the land management com- munity (Miles et al., 2006). A number of decision support tools created in the research domain might also be adapted for improved land manage- ment (e.g., verification of carbon sequestration projects). Finally, none of the planned CCSP synthesis and assessment products is focused on land use. An assessment of the impacts of predicted climate change on U.S. land use and strategies for adaptation could provide a useful tool to the policy community. Progress in all of these areas depends on sustained observations from Landsat-class satellites. With the failure of the scan line corrector on the Landsat 7 instrument in 2003, a data gap is now inevitable. The Landsat program has been the primary source of data for much of the research on land use and land cover change, providing the basis for quantifying local changes and trends at the regional scale and initiating land use models. No replacement instrument is available, and the proposed Landsat Data Continuity Mission will not be launched before 2011. A fully function- ing Landsat-class mission is a key part of a comprehensive land cover

76 EVALUATING PROGRESS OF THE U.S. CCSP monitoring system needed, for example, to quantify the rates of tropical deforestation and the fate of deforested land in different regions (Skole et al., 1997; Aspinall and Justice, 2004). In the meantime, the CCSP could facilitate an international initiative in the context of the Global Earth Ob- serving System of Systems on closing the gap with available foreign assets (Kintisch, 2007). Finally, because land cover and land cover change are basic measure- ments for international conventions and assessments (e.g., United Nations Framework Convention on Climate Change, Biodiversity Convention), considerable benefit could be gained from maintaining and strengthening links between the CCSP and international land cover-related programs, both to bring the results of the U.S. research program to the international community and to provide a forum for international coordination (DeFries et al., 2006). CARBON CYCLE The carbon cycle research element seeks to quantify the exchanges of carbon among the atmosphere, biosphere, and ocean and learn how these flows change as a result of human activity. A key element of the program is to determine the sign and magnitude of feedbacks between the carbon cycle and climate, especially those expected to operate in the coming century. This research is important for predicting future levels of carbon dioxide and methane in the atmosphere and the potential impacts of different carbon management strategies. Carbon observation networks will become increas- ingly important if governments begin to regulate carbon dioxide emissions, and understanding processes that might enhance or reduce carbon sinks will be required to design management strategies and judge their effects. The research strategy laid out in the CCSP strategic plan incorporates focused research and observation strategies developed by scientists to accel- erate progress in U.S. carbon cycle research (Sarmiento and Wofsy, 1999). Implementation of the CCSP has resulted in new, multiagency science pro- grams, including the North American Carbon Program (Wofsy and Har- riss, 2002) and the Ocean Carbon and Climate Change (OCCC) program (Doney et al., 2004), as well as enhanced efforts within agencies, such as the National Science Foundation’s (NSF’s) water and carbon in the Earth system competition, and a planned NASA satellite mission to monitor CO2 concentration in the atmosphere (Crisp et al., 2004). The program is guided by an active IWG and scientific steering committee that provides input on priorities and gaps in research plans.

PROGRESS TOWARD THE RESEARCH ELEMENTS 77 Progress Toward Answering the Research Questions As outlined below, good progress has been made toward answering CCSP research questions 7.1-7.4, although the overall objective of appor- tioning carbon dioxide emissions among atmospheric increases and land and ocean sinks has not been achieved. Research question 7.5 deals with understanding carbon cycle feedbacks for predicting future carbon dioxide and methane concentrations and is in an earlier stage of scientific enquiry. Although CO2 increases from fossil fuel burning and land use change are the clear result of human activities, links still have to be forged between ba- sic and social science, management, and policy to improve uncertainties in future carbon source scenarios (Dilling, 2007a, b). Question 7.6 addresses active carbon management strategies but is among the least developed of the carbon cycle efforts. We cannot yet account for the fate of the carbon we emit to the at- mosphere at the global or continental scale. Specific areas in which more research is needed to reduce uncertainties, such as carbon cycling in coastal oceans, have been identified through workshops (Doney and Glover, 2005) and in various carbon cycle planning documents (Wofsy and Harriss, 2002; Doney et al., 2004). North American Carbon Budget and Implications for the Global Carbon Cycle (synthesis and assessment product 2.2; CCSP, 2007b) is undergoing revision following its initial review and contains up- dated syntheses related to many of the carbon cycle research questions. Q 7.1. What are the magnitudes and distributions of North American car- bon sources and sinks on seasonal-to-centennial time scales, and what are the processes controlling their dynamics? The North American carbon budget is being formally assessed in syn- thesis and assessment product 2.2 (state of the carbon cycle), but improve- ments in observations and modeling approaches are required to reduce uncertainties. The approach toward answering question 7.1 is outlined in the North American Carbon Program (Wofsy and Harriss, 2002) and its implementation plan (Denning et al., 2005). The approach for understand- ing land carbon sources and sinks is (1) to scale up information from sur- face observational networks using models and remote sensing, and (2) to compare the resulting picture of the spatial distributions of carbon sources and sinks with those inferred from inversion of atmospheric CO2 anomalies using transport models. A number of recent syntheses demonstrating the utility of continuous data from observational networks are now available. Examples include the estimation of carbon sequestered in regrowing forests from the USDA’s For- est Inventory and Analysis Program (Goodale et al., 2002; Birdsey, 2006;

78 EVALUATING PROGRESS OF THE U.S. CCSP Birdsey et al., 2006), and direct measurements of land-atmosphere carbon exchange from the AmeriFlux network (Law et al., 2002; Hollinger et al., 2004). Fair progress has also been made in assimilating surface flux data into models and regional assessment of the impact of changing weather (e.g., European drought; see Ciais et al., 2005). However, the existing AmeriFlux network is not currently able to operate as the “integrated, near- real time network” envisioned to support the goals of the NACP (Wofsy and Harriss, 2002). “Top-down” inversion approaches take advantage of monitoring of atmospheric CO2 and methane mixing ratios (NOAA Earth System Research Laboratory, Global Monitoring Division; Dlugokencky et al., 2003) and have been tested at the regional scale using aircraft data (Lin et al., 2006). Tracers such as CH4, CO, isotopes of carbon, and SF6 (see Denning et al., 2005) and O2-N2 (Manning and Keeling, 2006), which emphasize the importance of specific sources or land cover types, provide additional observational constraints. Both approaches will be tested in the ongoing midcontinental intensive field campaign (Denning et al., 2005). However, planned expansion of tall towers and improved calibration at AmeriFlux towers to support experiment have not been fully realized as outlined in the NACP report or implementation plan. Good progress has been made in understanding terrestrial ecosystem processes, including determining the complex reasons underlying interan- nual variation in storage of carbon by forest stands (e.g., Barford et al. 2001) and understanding the importance of disturbance in determining ecosystem-atmosphere CO2 fluxes (Rapalee et al., 1998; Law et al., 2004; Saleska et al., 2003). Synthesis of several ongoing Free Air CO2 Enrichment experiments (King et al., 2004; Norby et al., 2005) showed similarities in response of vegetation across a number of forested ecosystems. Other experiments have used experimental manipulations to elucidate effects of warming, tropospheric O3, nitrogen deposition, and drought on ecosystem carbon storage and dynamics. While there is growing recognition of the importance of processes acting at decadal and longer time scales (e.g., disturbance, response to land management change), some key feedbacks that operate on those time scales are unquantified, including whether the decomposition of more stable forms of soil organic matter will accelerate with warming and feedbacks between warming, drought, and fire in the tropics (Cox et al., 2000). Although fair progress has been made toward comparison of bot- tom-up and top-down prediction approaches (Pacala et al., 2001), major uncertainties remain because of the relatively sparse nature of stations in the observation network, uncertainties associated with the transport models (Baker et al., 2006), and problems with the resolution of meteorological data needed for inversions (Denning et al., 2005). A number of modeling advances have been made that use model data fusion or data assimilation

PROGRESS TOWARD THE RESEARCH ELEMENTS 79 approaches to improve parameterizations of predictive process models (Denning et al., 2005). Although progress has been good in improving predictions of the spatial distribution of fossil fuel sources (Denning et al., 2005), predictive understanding of how sources in urban and suburban areas will evolve over time is lacking (Pataki et al., 2006). Good progress has been made in synthesizing and assessing existing data, as exemplified by CCSP synthesis and assessment product 2.2 (CCSP, 2007b). Risk management and decision support efforts are in their early stages (see question 7.6 below). Efforts related to the NACP include build- ing partnerships and linking observational networks in Mexico, Canada, and the United States. A few studies have linked agricultural practice to carbon sequestration potential (Lal et al., 2003), and the USDA undertakes greenhouse gas assessments (USDA, 2004). Q 7.2. What are the magnitudes and distributions of ocean carbon sources and sinks on seasonal to centennial time scales, and what are the processes controlling their dynamics? Focused research efforts and the synthesis of decades of observations have reduced uncertainties about the size of the ocean carbon sink, but significant uncertainties on ocean carbon processes remain. A science plan for the Ocean Carbon and Climate Change program (Doney et al., 2004) summarizes much of the recent progress and remaining uncertainties and suggests strategies for future research. The North American component of that research effort is also described in Denning et al. (2005). The proposed OCCC global observation network involves repeat hy- drographic surveys, remote sensing, and a new North American coastal observing network (Doney et al., 2004). Good progress has been made in narrowing uncertainty in the magnitude of the historical ocean carbon sink based on analyses of repeat hydrography measurements (Sabine et al., 2004) and in testing that estimate with observations of the atmospheric O2-N2 ratio (Manning and Keeling, 2006). However, a lack of observations still means that other estimates of air-sea CO2 exchange based on surface CO2 and gas exchange remain highly uncertain, especially with respect to interannual variations (Doney et al., 2004). Although satellite observations of ocean color (Sea-viewing Wide Field-of-view Sensor [SeaWiFS], MODIS) have enabled global mapping of ocean biota and assessment of interannual variations, problems with calibration and removal of instruments from NPOESS threaten the future continuity of ocean color records. Technology development (e.g., CO2 sensors that can make continuous measurements on buoys) has made good progress, and further developments are envisioned as part of the OCCC program. Good progress in understanding and representing processes has been

80 EVALUATING PROGRESS OF THE U.S. CCSP made on several fronts. The potential importance of ocean acidification in terms of ecological effects on calcifying organisms and a potential feedback between reduced calcification and atmospheric CO2 have been recognized (Feely et al., 2004). Programs such as the Joint Global Ocean Flux Study (Fasham, 2003) have yielded new insights into the functioning of ocean ecosystems and the efficiency of the “biological pump” that transports carbon from the surface to the deep ocean, although interannual variations in these processes (e.g., responding to ENSO variability) can be assessed at only a few stations (Doney et al., 2004). As is the case for terrestrial systems, slower, decadal-scale processes that operate in the oceans remain poorly understood. Important areas of uncertainty remain, including ocean- atmosphere gas exchange rates in the Southern Ocean (Ho et al., 2006) and quantification of the role of coastal oceans (Doney and Glover, 2005). Progress toward coupled ocean-climate models has been good, but fundamental uncertainties limit their predictive capability. For example, models generally predict reduced uptake of anthropogenic CO2 by the oceans based on decreased solubility and increased ocean stratification (Doney et al., 2004). However, the degree to which this might be offset by a reduction in export of carbon to the deep sea by marine biota is highly uncertain. Other major uncertainties include the degree to which ocean ecosystems will respond to altered dust inputs and the effects of warming on thermohaline circulation. Inadequate progress has been made in risk management and decision support. A number of mitigation strategies involving the ocean have been proposed, including direct injection of CO2 into the deep ocean and fertil- ization of the ocean biosphere with iron. Although fair progress has been made toward assessing the constraints of these two strategies, the scientific basis for fully assessing the processes involved and the potential conse- quences of such management activities is still in the developmental stages (Doney et al., 2004). Q 7.3. What are the effects on carbon sources and sinks of past, present, and future land use change and resource management practices at local, regional, and global scales? Good progress has been made in understanding the historical relation- ship between land use and the carbon balance, but great uncertainties in future land management scenarios limit predictive capability. Determining the extent to which land use change affects carbon sources and sinks re- quires two pieces of information: a record of land use change in the past, and an accounting of how different land use practices lead to carbon stor- age or loss. The first requires coordination with the land use and land cover research element (see “Land Use and Land Cover Change” above). The

PROGRESS TOWARD THE RESEARCH ELEMENTS 81 second requires observations and process studies to determine the relation- ship between carbon sources and sinks and the evolving state of land cover. A recent synthesis (Houghton, 2003) documented the sources of current uncertainties at the regional to global scale, which are a particular issue for estimates of the CO2 emissions from tropical deforestation (Hirsch et al., 2004). Good progress has been made within agency programs, especially in reconstructing the effects of past land use change on the North American carbon balance (Goodale et al., 2002; Birdsey, 2006; Birdsey et al., 2006), conservation reserves (Follett et al., 2001), and agricultural lands (Johnson et al., 2005; Martens et al., 2005). However, efforts to measure the net radiative (net greenhouse gas balance plus surface energy balance change) effects of land cover change have not always been well coordinated with these efforts (Randerson et al., 2006). Answers to Q7.3 at the local scale are required to inform management strategies involving future land use, but little research has been translated to information that might be useful for management decisions, especially those potentially involving credits for carbon sequestration (IPCC, 2000). Q 7.4. How do global terrestrial, oceanic, and atmospheric carbon sources and sinks change on seasonal-to-centennial time scales, and how can this knowledge be integrated to quantify and explain annual global carbon budgets? Although fair progress has been made in linking changes in regional and global rates of CO2 accumulation to climatic anomalies such as ENSO, understanding of the processes underlying some of these relationships is poor and limits our ability to predict factors that will dominate in the future. The same issues raised in questions 7.1 and 7.2—especially with respect to predictability, synthesis and assessment, and decision support— apply to this research question. Global-scale efforts rely on inversion of atmospheric observations using transport models and require coordination of the various trace gas monitoring networks. A planned satellite to map column CO2 inventory globally (Orbiting Carbon Observatory) will require testing with in situ data, and perhaps augmentation of existing atmospheric sampling networks. The information added by observations of gases other than CO2 (e.g., O2, CO, CFCs, methane, isotopes of CO2 and methane) aids in the attribution of sources and sinks globally (e.g., Bousquet et al., 2006; Manning and Keeling, 2006). Transport models and the sparse den- sity of air sampling networks currently limit this approach (Baker et al., 2006). However, recent work combining remote sensing of fire and trace gas observations highlighted the importance of fire associated with tropical deforestation to interannual variation in land-atmosphere CO2 exchange (van der Werf et al., 2006). A recent summary of ocean observations (Feely

82 EVALUATING PROGRESS OF THE U.S. CCSP et al., 2006) focused on changes in ocean-atmosphere trace gas fluxes as- sociated with interannual variations in tropical ocean upwelling and wind patterns (e.g., ENSO). Maintaining good progress on data will require strong international coordination among observation networks. Q 7.5. What will be the future atmospheric concentrations of carbon diox- ide, methane, and other carbon-containing greenhouse gases, and how will terrestrial and marine carbon sources and sinks change in the future? Predictions of future fossil fuel CO2 emissions as well as carbon sources and sinks associated with future changes in land management (the two largest uncertainties in the future carbon budget) are limited by the lack of involvement of stakeholder communities. Although good progress has been made in implementing coupled carbon-climate models, poor process understanding—especially of the magnitude (and sign) of feedbacks with climate—severely limits prediction of the future carbon balance of unman- aged lands and the ocean carbon balance. Improving predictions of future carbon dioxide levels requires not only an understanding of land and ocean carbon feedbacks to climate as incorporated in coupled models, but also improved ability to predict carbon sources from fossil fuel and land cover change based on human behavior. A community effort to build models with predictive capability into the next century that specifically couple carbon cycle and climate models is under way and provided input to the recent IPCC report (Fung et al., 2005). These model experiments show that the land and oceans decrease their capacity to act as repositories of fossil fuel CO2 in a future with higher global temperatures and fossil fuel emissions (Fung et al. 2005). Although these models are a major achievement, their predictive capability is limited in several areas. First, although records of the amount of past fossil fuel burning are updated regularly, and the spatial and temporal resolution of emissions has been improved (Marland et al., 2005), scenarios of future emissions require updating, especially in light of the sometimes very local scale of planning and carbon management strate- gies. Second, a number of uncertainties are associated with future land cover and land use changes and how they will influence whether the land is a net carbon source or sink. Human behaviors that will drive land use change (particularly the deforestation source), as well as improved under- standing of how (and how fast) ecosystems can adjust to changing climate (e.g., lengthened growing season, melting of permafrost), have not yet been incorporated into predictive models. Other sources of uncertainty are associated with our limited under- standing of the processes that influence the magnitude and sign of ocean and land feedbacks between carbon and climate (Sarmiento and Gruber, 2002; Fung et al., 2005). For example, neither the feedback between temperature,

PROGRESS TOWARD THE RESEARCH ELEMENTS 83 organic matter decomposition rates, nor the sign of carbon exchange be- tween land and atmosphere over the next century are well understood, and uncertainties remain about how temperature sensitivities scale from short (less than a season) to long (decades) time scales (Davidson and Janssens, 2006). Similarly, potential feedbacks between climate and ocean circulation may affect the net oceans CO2 exchange, but these processes are not yet understood in sufficient detail to be included in models that predict the next century. Overall, the development of better models will require progress in answering questions 7.1-7.5 above. Q 7.6. How will the Earth system, and its different components, respond to various options for managing carbon in the environment, and what sci- entific information is needed for evaluating these options? Informing carbon management is a new area of emphasis for the carbon cycle research element, so inadequate progress has been made on this re- search question. Efforts to quantify carbon changes that might accompany a given management practice have only recently begun (Dilling, 2007a, b). One notable exception is in the area of agricultural (Johnson et al., 2005; Martens et al., 2005) and forest (Birdsey et al., 2006) management. A stakeholder workshop was held in November 2004 as input to CCSP synthesis and assessment product 2.2, but potential stakeholders and the kinds of useful products that CCSP research could produce are still being identified (Dilling, 2007a, b). Future progress will depend on improvements in these areas, as well as coordination with the Climate Change Technology Program (e.g., alternative energy implications for the atmosphere) and the ecosystems and land use and land cover research elements. Opportunities and Threats Although good progress has been made toward balancing the global carbon budget, the fate of a portion of the CO2 added to the atmosphere by fossil fuel burning and deforestation remains unresolved. In part, this is because the strengths of ocean and land carbon sinks can change from one year to the next, making it difficult to patch together a coherent set of observations at the proper temporal and spatial scales for interpretation. Observations take place on short time scales of minutes to several years, whereas processes that operate on longer time scales (e.g., erosion, deposi- tion, vegetation mortality and regrowth, ocean circulation changes) set the stage for these short-term fluxes. Understanding of the magnitude and even the sign of feedbacks to climate or land use of these longer-term processes   See <http://cdiac.ornl.gov/SOCCR/workshop1.html>.

84 EVALUATING PROGRESS OF THE U.S. CCSP is uncertain. For example, North America land carbon sinks reflect the dynamics of forest regrowth, fire suppression, or enhancement of plant growth by CO2 fertilization (e.g., Pacala et al., 2001). Similarly, ocean carbon flux estimates in the future are limited by our understanding of how ocean thermohaline circulation and biology may change with climate (Sarmiento and Gruber, 2002). The main factor limiting progress is not our understanding of where key uncertainties lie, but our ability to carry out the research needed to reduce those uncertainties under current budget constraints. For example, inversions of observations of gradients in atmospheric CO2 (and other trace gases) can in theory provide measures of the locations of major carbon sources and sinks. However, making reliable estimates requires (1) invest- ment in trace gas transport models, (2) expansion of networks to increase data density and to measure the free troposphere in addition to the bound- ary layer, and (3) expansion of a suite of trace gas and isotope measures. ECOSYSTEMS Ecosystems “supply food, fiber, fuel, clean air and water and many other goods to society” (CCSP, 2006a). By altering the structure and func- tion of marine and terrestrial ecosystems, climate change potentially af- fects all aspects of society. The CCSP ecosystem research element aims to provide a scientific basis for the development of policies and procedures that will protect the goods and services derived from marine and terrestrial ecosystems. Progress Toward Answering the Research Questions Progress has been made in addressing aspects of all the overarching research questions that guide the ecosystem research element. For example, plans and strategies for managing some marine and terrestrial ecosystems affected by climate change have been developed (e.g., Dale et al., 2001; Mu- rawski and Matlock, 2006). Efforts to couple climate and ecosystem models are established (e.g., Cramer et al., 2001; Schmittner, 2005). However, much remains to be done before the effects of climate change on marine and terrestrial ecosystems can be predicted (Burkett et al., 2005). Contin- ued progress will require long-term support for data systems, development of integrated modeling frameworks with data assimilation and predictive capability, and continued refinement of our understanding of processes, feedbacks, and linkages within ecosystems and between ecosystems and the larger Earth system.

PROGRESS TOWARD THE RESEARCH ELEMENTS 85 Q 8.1. What are the most important feedbacks between ecological systems and global change (especially climate), and what are their quantitative relationships? Progress toward identifying feedbacks between ecological systems and global change and describing quantitative relationships for these linkages has been inadequate. Most of the research has been directed at understand- ing changes that will occur in ecosystems as a result of climate change (see question 8.2). Understanding feedbacks and quantitative relationships requires integrated modeling coupled with coordinated data collection pro- grams that sample multiple temporal and spatial scales. However, progress on defining the types of measurements and models that are needed to ad- dress this research question for terrestrial (Hurtt et al., 1998) and marine (Doney et al., 2003) ecosystems has been inadequate. Progress will also depend on results of research being undertaken under the carbon cycle re- search element and on long-term in situ and satellite-based remote sensing capabilities. Current capabilities to study and monitor coastal ocean ecosys- tems are limited, and the planned high-resolution coastal water imager on the Geostationary Operational Environmental Satellite Series R (GOES-R) has been cancelled. Q 8.2. What are the potential consequences of global change for ecological systems? The bulk of research carried out within the ecosystems research ele- ment falls under this research question, and good progress has been made in answering it. The LTER program, which includes sites on both man- aged (e.g., farmland) and less managed landscapes (e.g., arctic tundra), has documented long-term (multidecadal) changes in terrestrial ecosystems in response to climate change and enabled process studies of the physical, chemical, and biological components of ecosystems. The recent expansion of the LTER network to include more marine sites should enhance our ability to study the effects of climate change on a variety of ecosystems. The developing National Ecological Observatory Network is extending such studies to continental scales and advancing understanding of how ecosystems and organisms respond to variations in climate. The Global Ocean Ecosystems Dynamics Program (Fogarty and Powell, 2002) and the Throughfall Displacement Experiment (Hanson and Wullschleger, 2003) led to improved documentation of the response of marine and terrestrial eco- systems, respectively, to the effects of climate variability. These and other programs have led to increased understanding of the potential consequences   Contributions from individual LTER sites can be found at <http://www.lternet.edu>.

86 EVALUATING PROGRESS OF THE U.S. CCSP of climate change for ecosystems, and attempts to use this information to guide ecosystem-based management of resources are beginning (e.g., see the June 2004 theme section of Marine Ecology Progress Series). Q 8.3. What are the options for sustaining and improving ecological sys- tems and related goods and services, given projected global changes? Progress on this research question has been good. Regulatory and management options and/or plans have been developed (and are under development), and the infrastructure is now in place to guide management of changing terrestrial and marine ecosystems. The plans employ an adap- tive approach to managing ecosystems, including agricultural and forest systems, that is intended to add resilience to the ecosystem (e.g., see USDA Northwest Forest Plan; Butler and Koontz, 2005; Bormann et al., 2007). Governance systems specific to ecosystems affected by climate change have also been developed. For example, the National Marine Fisheries Service uses plans that link fisheries management to climate variability (Murawski and Matlock, 2006). An important factor in the progress of this research question is identification of the relevant stakeholders and involvement of these individuals in ecosystem management. Studies are under way to identify indices, bioindicators, and biocrite- ria that can be used to describe the current state of marine and terrestrial ecosystems and to project the consequences of climate change for these systems. Much of the focus thus far has been on managed ecosystems, but unmanaged ecosystems are equally important in understanding responses to climate change and also deserve attention. Opportunities and Threats The temporal and spatial scales of ecosystem change that are important for management need to be better defined. Adaptive management structures are in place for managed ecosystems, but the consequences and feedbacks from the actions dictated by these structures have not been explored (e.g., Everglades example, Gunderson and Light, 2006). For unmanaged ecosys- tems, no clear stakeholder exists, but neglecting these large and possibly important ecosystems may result in unwanted surprises and unanticipated feedbacks. Understanding climate change effects in managed and unman- aged ecosystems is important to sequestration technologies that do not consider a priori potential effects on ecosystem structure and function.   See also materials on the Ecosystem-Based Management Tools Network, <http://www. embtools.org>.   Projects and publications appear at <http://www.epa.gov/bioindicators/coral/index.html>.

PROGRESS TOWARD THE RESEARCH ELEMENTS 87 Improvements in understanding linkages between atmospheric, ecosystem, and water cycle processes are needed for terrestrial and marine ecosystems. However, current funding structures do not generally encourage research programs that span a wide range of potential interactions and inputs. Con- sequently, most research programs focus on one or two aspects of these interactions, which provides only limited insights into possible effects of climate change. Areas in which future investments in ecosystem research can result in significant advances in understanding are (1) improving atmospheric trans- port models to better take advantage of existing and planned observational capabilities and coupling these models to marine and/or terrestrial ecosys- tem models; (2) supporting development of ecosystem models that include inputs of reanalysis products, forward models, and data assimilative models that are at the cutting edge of model development; (3) supporting the de- velopment of ecosystem models and coupled models (e.g., atmosphere-eco- system, ocean circulation-ecosystem) that include the effects of disturbances and long-term changes; and (4) developing monitoring and observations systems that can document the impacts of changing ecosystems on humans and feedbacks to climate. Ecosystems are already beginning to respond to climate change, and it is imperative to develop the modeling and research infrastructure to predict the possible outcomes and consequences. Continued progress in the characterization and understanding of eco- system change and its consequences depends on the availability of long- term observations of atmospheric, terrestrial, and marine systems. Existing observational networks, such as the TAO array in the tropical Pacific Ocean, continue to be important and to need ongoing maintenance, and new remote sensing (e.g., GOES-R satellite) and in situ observing capabili- ties would improve understanding of ecosystems in the coastal ocean. Also, intercalibration of legacy and operational observing systems (e.g., MODIS, SeaWiFS, Coastal Zone Color Scanner) would provide a basis for assessing long-term (multidecadal) effects of climate change. Finally, ecosystems and the carbon cycle are closely linked; there are many different and possibly overlapping feedbacks (e.g., changes in meth- ane). Yet most current research programs in the agencies and the CCSP as a whole consider them separately. Progress in both research elements could be fostered by a more coordinated approach. HUMAN CONTRIBUTIONS AND RESPONSES TO ENVIRONMENTAL CHANGE The CCSP currently manages research on human dimensions, decision support tools, and human health effects of climate change together. These areas are interconnected, but distinct. Human dimensions research involves

88 EVALUATING PROGRESS OF THE U.S. CCSP a very broad set of research questions and disciplines. The topics are some of the most fundamental in the arena of climate change as an environmental problem (as distinct from an interesting scientific puzzle), including how humans affect climate processes; how societies’ and people’s well-being is affected (positively and negatively) by changes in climate and by actions taken to mitigate or abate the effects of climate change; and how societies respond, cope, and adapt to climate-related impacts. The disciplines in- volved in the human contributions and responses research element include demography, psychology, geography and regional sciences, economics, an- thropology, political science, and sociology (CCSP, 2003). Decision support includes research on ways to get climate information used in decision making, the development of tools, and other activities similar to those traditionally associated with extension functions. It also includes research on and application of a systems engineering approach to decision making as exemplified by NASA’s program focusing on the use of data generated by its Earth Observation System in decision making. Although decision support activities often draw on results from human dimensions research, the latter is broader in scope and includes basic social sciences to understand and explain both anthropogenic causes of climate change and potential consequences of climate change for societies, cultures, political systems, and individuals. For example, research on how individu- als make decisions under great uncertainty will clearly have payoffs in the decision support arena. NSF’s program on Decision Making Under Uncer- tainty (DMUU), which has established five university centers, is a promising example of how human dimensions resources can be used to produce both basic and decision-driven science of great relevance (CCSP, 2007a; McNie et al., 2007; Sarewitz and Pielke, 2007). Finally, health effects research, as defined in the CCSP strategic plan, includes data collection, studies to understand potential effects of global environmental change on health, and assessment of the cumulative risk of negative effects of climate and environmental change on human health. A single interagency working group handles all three topics, and prog- ress and future plans for the three are reported together in Our Changing Planet. Combining management of human dimensions research and decision support tools deemphasizes the need for basic research in the social sciences throughout all the CCSP overarching goals. Moreover, the inclusion of re- search on the effects of ozone on health and systems engineering aspects of decision support resources in the budget makes it harder to determine the amount of resources being invested in human dimensions research. Con- sequently, to evaluate progress in the human contributions and responses research element, the committee had to obtain separate programmatic and budget information from the CCSP (see Appendix B). Research questions for the CCSP human contributions and responses

PROGRESS TOWARD THE RESEARCH ELEMENTS 89 research element encompass the main areas of inquiry, including determin- ing the causes and consequences of human drivers of global climate change; understanding impacts and differential levels of vulnerability and adaptive capacity; and developing methods and capacities to improve societal deci- sion making under conditions of uncertainty and complexity. One of the questions also concerns understanding the human health effects of global climate change. Progress Toward Answering the Research Questions Important research in human dimensions has been carried out by a committed, if small, research community, despite the modest investment research thus far (about $25 million to $30 million per year; Appendix B). Significant findings have been published on both the human causes of global climate change and its impacts on societal well-being in the United States and other countries. In addition, a substantial portion of this research has been stakeholder driven and has resulted in positive interactions across the science-society divide, which not only created opportunities for decision- relevant research but also enhanced our understanding of opportunities and constraints for CCSP science-generated knowledge to affect decision making. The research on human dimensions appears to be of high quality, particularly work undertaken as part of NSF programs (e.g., DMUU cen- ters, Harvard knowledge systems for sustainable development project; see Cash, 2001; Cash et al., 2006; Clark and Holliday, 2006; van Kerkhoff and Lebel, 2006) and DOE’s program on integrated assessment modeling. The DOE program has coupled long-term support for major research programs at the Massachusetts Institute of Technology (MIT) and the Joint Global Change Research Institute (Pacific Northwest National Laboratory) with a diverse portfolio of smaller-scale research programs that focus on how natural science, economics, and other social science are integrated into policy models for climate change. However, many research gaps remain, and both the size of the human dimensions community and the level of available funding seem inadequate to carry out the research necessary to answer all of the research questions. Q 9.1. What are the magnitudes, interrelationships, and significance of pri- mary human drivers of and their potential impact on global environmental change? Progress in answering this research question has been inadequate. Our Changing Planet reports two projects that focus on the dynamics of human drivers of climate change (CCSP, 2005b). One study examined the relation- ship between income and the use of traditional fuels (e.g., firewood) versus

90 EVALUATING PROGRESS OF THE U.S. CCSP commercial fuels for home heating and cooking in rural China, and the other study examined the role of household demography in decisions on land use, especially deforestation. Some research on human drivers has also been conducted outside the CCSP. However, synthesis and integration of results across human dimensions disciplines has been limited. For example, greenhouse gas emission scenarios continue to be based on simple models involving a few drivers (e.g., population, affluence, technological change). Recent studies are beginning to explore how these drivers affect each other and how they interact with other major social changes (e.g., urbanization, industrialization) and with environmental factors (e.g., tropical or temper- ate location) (York et al., 2003a, b; Rosa et al., 2004). Current understanding of the effects of human drivers on ecosystem change and, in turn, the effects of changes in ecosystem services on human well being is meager (Millennium Ecosystem Assessment, 2006). Changes in ecosystem services, including those caused by climate variability, are al- most always due to multiple, interacting drivers that work over time. These changes operate over multiple temporal, spatial, and governance scales and can also feed back to drivers. No existing conceptual framework captures the broad array of findings from the large bank of case studies presented in the Millennium Ecosystem Assessment. Q 9.2. What are the current and potential future impacts of global environ- mental variability and change on human welfare? What factors influence the capacity of human societies to respond to change and how can resilience be increased and vulnerability reduced? A few lines of research have shown promise, but considerably more effort and resources have to be expended to begin to answer this question. Although RISAs focus on climate variability and change, these regionally based programs have (1) produced valuable insights on institutional op- portunities and constraints on the use of climate knowledge by decision makers in different application sectors (e.g., water resources, fire and risk management, agriculture); (2) assessed vulnerabilities of a few groups of stakeholders; and (3) developed innovative methodologies to understand and manage the interaction between scientists and stakeholders (McNie et al., 2007). In addition, NOAA-sponsored research on the economics and human dimensions of climate variability and change has identified poten- tial impacts of climate-related phenomena on different sectors (e.g., water, agriculture, coastal areas). The same program has also sponsored a few projects focusing on vulnerability assessment and adaptation. Although a   See project descriptions and a list of publications at <http://www.climate.noaa.gov/index. jsp?pg=./cpo_pa/cpo_pa_index.jsp&pa=sarp&sub=3>.

PROGRESS TOWARD THE RESEARCH ELEMENTS 91 substantial portion of this research focused on climate variability, its find- ings have relevance to the transfer and diffusion of climate information to decision makers in different sectors working at smaller scales. A significant part of this research is being reported in synthesis and assessment product 5.3 (see Appendix A). Finally, a few assessments of vulnerability have been sponsored by DOE (Moss et al., 2001) and NSF (e.g., vulnerability of coastal communities; see Appendix B). However, these projects are minuscule relative to the magnitude of the question. Much more research is needed, especially in understanding the impacts of and adaptation to climate change across different sectors and geographical regions, mapping differential vulnerabilities, and designing interventions to build resilience. Similarly, progress on the economics of climate change has generally been inadequate, although a recent U.K. report was an important contribution to the field (Stern, 2007). Q 9.3. How can the methods and capabilities for societal decision making under conditions of complexity and uncertainty about global environmental variability and change be enhanced? Overall, progress in advancing capabilities for decision making has been inadequate, but some significant research has been carried out to char- acterize uncertainty and complexity in the context of global climate change, to understand their impact on decision making and management, and to understand the links between producers and users of climate science. Four programs stand out as successes: DMUU centers, RISAs, DOE’s Integrated Assessment Program, and the Harvard knowledge systems project. Within the RISA programs, for example, some original data on potential impacts and governance responses (from both the public and the private sector) have been generated (e.g., Callahan et al., 1999; Hartmann et al., 2002; Pagano et al., 2002; Carbone and Dow, 2005; Jacobs et al., 2005; Lemos and Morehouse, 2005; O’Connor et al., 2005; CCSP, 2007a). However, RISA-generated data are mostly at the regional level and limited to sectors relevant at this scale, such as water in California or fisheries in the Pacific Northwest. Each of these four programs has made fair progress in understanding and characterizing uncertainties related to both physical and institutional processes affecting and being affected by global climate change. Some stud- ies have addressed the need to incorporate information from climate science into decision making and how to evaluate predictability and predictive capabilities of different physical and socioeconomic models, but this work is at an early stage. Finally, they have assessed and synthesized knowledge in their focus areas (e.g., Cash et al., 2003; CCSP, 2007a; McNie et al., 2007). In addition, DOE’s long-standing support of major integrated assess-

92 EVALUATING PROGRESS OF THE U.S. CCSP ment projects has led to increased capabilities to conduct these assessments; major modeling teams at the Joint Global Change Research Institute and MIT, as well as a number of other researchers, are now working in this area. Some of this work is related to decision support and some to human dimensions research. However, the total output from these efforts has been low for the complexity and high levels of uncertainty that still characterize the physical processes causing global climate change and the magnitude of the potential impacts on socioeconomic and ecosystems (e.g., Millennium Environmental Assessment, 2006; Stern, 2007). Q 9.4. What are the potential human health effects of global environmental change, and what climate, socioeconomic, and environmental information is needed to assess the cumulative risk to health from these effects? The vast bulk of this research program involves either health effects of ultraviolet radiation or satellite measurement of particulate matter con- centrations for health-related analysis. A few research projects focusing on the intersection of climate, health, and human dimensions have been car- ried out under the auspices of the Environmental Protection Agency and the Centers for Disease Control and Prevention (see Appendix B; CCSP, 2005b, pp. 131-132). For example, a health impacts assessment (Patz et al., 2000) examined the interactions between health and climate variability and change, and identified adaptation strategies. Opportunities and Threats A review of the CCSP strategic plan recommended accelerating efforts in human dimensions, economics, adaptation, and mitigation by strength- ening science plans and institutional support (NRC, 2004c). The inadequate progress of the human contributions and responses research element may reflect organizational problems within the agencies and the CCSP. Of par- ticular concern are the absence of social science leadership to guide the program and sufficient resources (dollars and people) to carry it out (see Table 2.1). Few agencies have programs dedicated to human contributions and responses, and CCSP funding devoted especially to human dimensions is significantly less than funding devoted to most of the other research ele- ments (Table 1.1). Human capacity may also be insufficient to carry out this work. The natural sciences may offer a successful model for building human dimensions capacity, especially programs to move young investigators into the arena and to support postdocs. The program could benefit from improved linkages to other programs, such as NSF’s biocomplexity program. Integration and enhanced support for human dimensions are especially critical given the potential for such

PROGRESS TOWARD THE RESEARCH ELEMENTS 93 research to inform decision making and the management of climate impacts on human, sociopolitical, and ecological systems. If the quality and “us- ability” of the few projects already funded are any indication, investment in human dimensions not only is necessary, but may also be highly cost effective. Improvement of existing data sets and the collection of new data at suitable resolution would also speed progress in human dimensions. A major need is for data sets on both climate-related human activities and environmental data at the same spatial and temporal coverage and resolu- tion. Many relevant social data sets exist at useful levels of aggregation, but they have not been geocoded or are not available in spatial forms that are readily linked to environmental data (e.g., they are coded by political juris- dictions rather than spatial coordinates). For example, DOE has collected energy consumption data on residential, commercial, and industrial users since the 1970s, but most available data are aggregated at only the state or regional level and cannot be used to model the drivers of greenhouse gas emissions at higher resolutions. Data on property values are collected by jurisdictions around the country and they appear on maps, but not in forms that facilitate linkage to climate models and thus estimates of the economic consequences of possible future floods or storms on particular places. The use of such data sets in models would enable projections of greenhouse gas emissions that are based on analyses of the driving forces and their inter- actions, rather than on simplified assumptions about a few driving forces. It would also provide an empirical base for disaggregated analyses of the human consequences of climate variability and change and of the potential benefits of various adaptive and mitigative responses. Finally, future evaluations of progress would be greatly facilitated if the CCSP reported accomplishments on human dimensions research separately from accomplishments on decision support activities and health effects research. PRELIMINARY ASSESSMENT OF THE OVERARCHING GOALS: AN EXAMPLE As noted in Chapter 2, it should be possible to use results of the pre- liminary evaluation of research questions to assess the overarching goals. An example of how the evaluation could be conducted for focus area 1.4 of overarching goal 1 is given below. The committee first mapped the research questions and relevant cross-cutting issues to the focus areas (Box 4.1). The mapping proved challenging because the connections are not all laid out in the CCSP strategic plan, and each focus area is connected to several research questions and often to one or more cross-cutting issues. The scores

94 EVALUATING PROGRESS OF THE U.S. CCSP BOX 4.1 Links Between Overarching Goal 1 Focus Areas, Research Questions, and Cross-Cutting Issues Overarching Goal 1: Extend knowledge of the Earth’s past and present climate and environment, including its natural variability, and improve understanding of the causes of observed variability and change Focus 1.1. Better understand natural long-term cycles in climate (e.g., Pacific Decadal Variability, North Atlantic Oscillation) Associated research questions: 4.2, 5.1, 8.2, and 9.2 Focus 1.2. Improve and harness the capability to forecast El Niño-La Niña and other seasonal-to-interannual cycles of variability Associated research questions: 4.2, 5.2, and 9.2 Focus 1.3. Sharpen understanding of climate extremes through improved ob- servations, analysis, and modeling, and determine whether any changes in their frequency or intensity lie outside the range of natural variability Associated research questions: 4.3, 4.4, and 8.2 Focus 1.4. Increase confidence in the understanding of how and why climate has changed  Associated research questions: 3.1, 3.2, 3.3, 4.4, 5.1, 5.2, 6.1, 6.2, 6.4, 7.1, 7.4, 8.1, and 9.1 Associated cross-cutting issues: 10.1, 10.2, and 10.3 (modeling) Focus 1.5. Expand observations and data and information system capabilities  Associated research questions: 3.1, 3.2, 3.3, 3.5, 4.1, 4.5, 5.2, 5.4, 6.1, 6.2, 6.4, 7.1, 7.4, 8.1, and 8.2  Associated cross-cutting issues: 12 (observing) and 13 (data management) subgoals and comments on the relevant questions were then combined to make an overall evaluation. Focus Area 1.4 The twentieth century has witnessed major changes in both climate forcing terms (greenhouse gases, aerosols, land use and land cover, volcanic emissions of SO2) and climate (e.g., surface temperatures, atmospheric tem- peratures, ice and snow cover, mountain glaciers). Understanding how and why these changes occur is important for evaluating the human impact on climate and predicting future changes. Thus, focus area 1.4 is a key com- ponent of the CCSP and involves several research questions. Focus area 1.4 is addressed by five research elements—including atmospheric composition, climate variability and change, water cycle, land use and land cover change, and human contributions and responses—and the modeling cross-cutting issue. It has two parts, which are evaluated separately below.

PROGRESS TOWARD THE RESEARCH ELEMENTS 95 How Has Climate Changed? Good progress has been made in this area. For example, the IPCC (2007) concludes that “warming of the climate system is unequivocal.” Research conducted under the CCSP, including the synthesis and assessment report on atmospheric temperature trends, played an important role in the IPCC’s finding. However, continued progress is seriously threatened by the loss of climate instruments on NPOESS and other satellites. The research part of this topic is covered under questions 4.2 and 4.4 of the climate variability and change research element. Sustained investments in observing systems and models have led to advances in understanding ocean processes and several natural forcing terms (e.g., solar insulation, volcanic emissions), as well as the relationship between climate variability and change, droughts, and wildfires. A limited number of paleoclimate records needed to advance understanding and improve predictions are also available. In addition, three synthesis and assessment products (1.1, 1.2, and 1.3) are relevant to this focus area. Synthesis and assessment product 1.1 (tem- perature trends) largely resolved the discrepancy between surface observa- tions of surface warming and satellite observations of atmospheric warming (CCSP, 2006b). The two other products have not yet been published (see Appendix A). Why Has Climate Changed? Major improvements have been made in quantifying the anthropogenic forcing terms (i.e., radiative forcing due to greenhouse gases, aerosol forc- ing, land use albedo forcing). For example, the IPCC (2007) concludes that the “globally averaged net effect of human activities since 1750 has been one of warming with a radiative forcing of +1.6 [+0.6 to +2.4] Wm–2.” However, large uncertainties remain in the magnitude of emissions of aero- sols, aerosol-cloud interactions, and the importance of tropospheric ozone forcing. Internal variability in the coupled land-ocean-atmosphere system, changes in natural climate forcing terms (solar insolation and volcanic emissions), and anthropogenic influences (changes in greenhouse gas emis- sions, aerosols, and land use and land cover) contribute to climate changes. Our understanding of these processes has improved significantly over the last few decades, fueled by the synthesis of different types of observations (satellite, aircraft, ship, buoy, land surface) and the integration of observa- tions and laboratory experiments. In addition, models (e.g., coupled ocean- atmosphere-land climate models, chemical transport models, carbon cycle models) have played a fundamental role in sorting out the various forcing factors that influenced the observed changes. Changes in climate forcing are covered in research questions 3.1, 3.2, 3.3, 6.1, 6.2, 7.1, 7.4, and 9.1. The

96 EVALUATING PROGRESS OF THE U.S. CCSP effects of climate change feedbacks on forcing are covered in questions 5.2, 6.4, and 8.1. Fundamental weaknesses still exist in the following areas: • Regional climate changes. Concerted efforts to quantify the impact of human activities on North American climate change and its subsequent consequences for agriculture, the water budget, and health have been lim- ited since the last major assessment of climate change impacts in the late 1990s (National Assessment Synthesis Team, 2000). • Role of cloud feedback in climate change. Changes in water vapor, clouds, and precipitation in response to changes in climate forcing and cli- mate change can have major feedback effects. Aerosol-cloud-precipitation feedbacks are also part of this issue. The effect of aerosols in inhibiting cloud formation has been measured, but large uncertainties remain about the emission sources of elemental and organic carbon, the indirect effect of aerosols on climate, and the importance of aerosol solar heating of the atmosphere on climate. • Feedback processes between the physical, chemical, and biologi- cal parts of the climate system. Progress in understanding these feedback processes, which may also have influenced the observed changes, has been inadequate. • Climate-societal interactions. Progress has been inadequate in the development of a quantitative understanding of how societal behavior and choices affect the environment and how societies in turn are affected by the environment. Good progress has been made in mapping land cover change, but these studies have been limited by difficulty in obtaining relevant socioeconomic data. With the exception of advances in land use change and decision mak- ing under uncertainty, inadequate progress has been made on understanding the human drivers of climate change. Ecosystems influence atmospheric composition of greenhouse gases, aerosol precursors, and absorption and reflection of solar radiation at the surface. However, research efforts to date have focused on understanding changes that will occur in ecosystems as a result of climate change. Scientific questions regarding the response of the climate system to nat- ural and anthropogenic forcing cannot be addressed with traditional physi- cal climate models (e.g., those that do not include interactive chemistry, the carbon cycle, or interactive aerosol models). Consequently, significant efforts have been made to extend physical models to include the interac- tions of climate with biogeochemistry, atmospheric chemistry, ecosystems, glaciers and ice sheets, and anthropogenic environmental change. The types of measurements and models needed to obtain a more comprehensive un- derstanding of feedbacks for terrestrial and marine ecosystems are still be-

PROGRESS TOWARD THE RESEARCH ELEMENTS 97 ing defined. Finally, U.S. underinvestment in computing power has limited progress in accurately representing key climate processes and feedbacks. Overall, the committee found that a fair amount of progress has been made on focus area 1.4. Slightly greater advances have been made in under- standing how climate has changed than why it has changed. These advances have been driven largely by the availability of a wide range of data from satellite and in situ networks, which have significantly improved our ability to represent physical quantities. Understanding of the forcing factors that affect climate—and vice versa—has progressed steadily, with the greatest gains in atmospheric composition and, to a lesser extent, the water cycle. Inadequate progress has been made in understanding ecosystem or human feedbacks and developing coupled models capable of addressing natural and anthropogenic forcing.

Next: 5 Progress Toward the Cross-Cutting Issues »
Evaluating Progress of the U.S. Climate Change Science Program: Methods and Preliminary Results Get This Book
×
Buy Paperback | $52.00 Buy Ebook | $41.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The U.S. Climate Change Science Program (CCSP) coordinates the efforts of 13 federal agencies to understand why climate is changing, to improve predictions about how it will change in the future, and to use that information to assess impacts on human systems and ecosystems and to better support decision making. Evaluating Progress of the U.S. Climate Change Science Program is the first review of the CCSP's progress since the program was established in 2002. It lays out a method for evaluating the CCSP, and uses that method to assess the strengths and weaknesses of the entire program and to identify areas where progress has not met expectations. The committee found that the program has made good progress in documenting and understanding temperature trends and related environmental changes on a global scale, as well as in understanding the influence of human activities on these observed changes. The ability to predict future climate changes also has improved, but efforts to understand the impacts of such changes on society and analyze mitigation and adaptation strategies are still relatively immature. The program also has not met expectations in supporting decision making, studying regional impacts, and communicating with a wider group of stakeholders.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!