UCAR > Communications > Staff Notes > May 2000 Search


May 2000

Assessing the climate: How it gets done

"Can't we say anything more than 'We can't say anything'?"

Jerry Meehl. (Photos by Carlye Calvin.)

CGD's Jerry Meehl recalls this plea from Tom Karl, head of the National Climatic Data Center, with a sympathetic smile. In his casual comment to Jerry at the 2000 meeting of the American Meteorological Society, Karl highlighted the tug of war at the heart of assessing climate change. Policymakers and the public want certainty. Atmospheric scientists can seldom provide it, but, as Karl realized, they have an obligation to say something. "In fact," says Jerry, "there are results that are more certain than others, and this needs to be communicated."

The next year is a big one for climate assessments. The U.S. National Assessment, the first of its kind, will be released this summer. Following in early 2001 will be the third global assessment from the Intergovernmental Panel on Climate Change (IPCC), which involves 2,500 scientists from around the world. Both analyses draw from NCAR-based work, and a handful of scientists at NCAR and UCAR are racking up thousands of in-flight miles and hours of sleep lag as they guide key portions of the reports to completion.

How we got here from there

The U.S. Department of Energy began summarizing climate-change research with its "State of the Art" reports in the mid-1980s. The IPCC era of international climate assessment began in 1988 when the World Meteorological Organization and the United Nations Environment Program joined forces. The WMO and UNEP wanted to go a step beyond earlier, individual reviews and find out what the best minds in the business thought collectively. These assessments would come to definitive conclusions--scientists' best estimates of what actually could happen--as opposed to earlier reviews, which simply summarized results.

What's emerged is "the biggest scientific assessment of all time," according to the journal Science. It's also a novel type of effort, notes Jerry. "The process itself is pretty remarkable. Do you know of any other discipline that tries to determine, every five years, the state of human knowledge?"

The first IPCC report, issued in 1990 and updated in 1992, suggested a plausible global temperature climb of 0.2 to 0.5°C (0.4-0.9°F) per decade over the next century due to increases in atmospheric carbon dioxide, but stopped short of attributing the 20th-century temperature rise to human factors. The 1995 report focused on the lower end of the expected temperature range by including the effects of aerosols, but its biggest impact came from a single phrase: ". . . the balance of evidence suggests that there is a discernible human influence on global climate."

Kevin Trenberth.

Critics of the 1995 report charged, among other things, that the participants had changed the models since 1990 to produce results that fit a preconceived notion. To Kevin Trenberth, CGD senior scientist and an IPCC veteran since the beginning, this accusation reflects a basic misunderstanding of what the reports have to offer. "The changes from 1990 to 1995 were in the scenarios, not the projections. Just about all that the IPCC has ever produced [is] scenarios: if we put in this set of emissions or concentrations, what will the climate change be?" The IPCC's core science section for 2000 will incorporate fairly recent findings from about 20 global climate models from around the world, including CGD's NSF-funded Climate System Model and DOE-funded Parallel Climate Model (soon to be merged into the new Community Climate System Model).

The upcoming report will provide an even broader range of scenarios for how climate might evolve and affect humans, and it'll explain how those scenarios were produced. This is as it should be, says Linda Mearns (ESIG). "There are a lot of choices you have to make when you construct a scenario," she explains. "Do you include just the mean changes, for example? Or do you also include some aspects of variability [over time]?" It was one of her concerted goals to get a chapter on scenario development into the report from IPCC's Working Group I. Linda is a convening lead author for that chapter, as well as a lead author for two others on using scenarios in impact assessment and on evaluating and projecting regional climate.

Linda Mearns.

The IPCC 2000 release, like the 1995 one, will be in three volumes. Working Group 1 deals with core science; WG2 addresses impacts, adaptation, and vulnerability; and WG3 tackles mitigation and related economic and social concerns. The biggest organizational change from 1995 is that the topic of mitigation (steps to reduce emissions) has been moved from WG2 to WG3. This allows adaptation to take a higher profile in WG2--perhaps a sign of newfound realism about the inevitability of some climate change.

A lot of work

"It's very time consuming," says Kathy Miller (ESIG) of the background work needed to contribute to IPCC. Kathy, a newcomer to the assessment, is serving as a lead author on chapter 15 of WG2, which focuses on North American climate impacts. "It's more work than it might appear on the surface, because the final product has to be so condensed. You have to read a phenomenal amount of material to be able to synthesize it." On the plus side, she says, "It's material you ought to be familiar with anyhow to make a credible contribution to your field. The range of material helps you put your own work in context."

Kathy Miller.

Along with Linda, there are two other convening lead authors at NCAR: Kevin (for a WG1 chapter on climate processes) and Jerry (for a WG1 chapter on projections of future climate change). Convening lead authors pull together the contributions from ten or so lead authors, and many more contributors, for each chapter. Kevin can testify that the task is no cake walk. Each chapter goes through a standard peer review and a politically oriented government review. The result can be hundreds of pages of comments. For Kevin's own chapter, which is running about 70 pages, "the comments were longer than the manuscript. All of them have to be dealt with." Furthermore, in a new step added for IPCC 2000, the response to these and later reviews is checked by one or more review editors for each chapter, who make sure that reviewers' concerns have been adequately addressed. This change was partially in response to controversy over last-minute comments and the small wording changes that followed late in the 1995 report process (see below).

The three main WG reports will be finalized over the next few months. For each report, highlights have been pulled together in a technical summary and a more lay-oriented policy summary; Kevin is part of the 20-member team producing both of these documents for WG1. Early next year comes the big enchilada: the intergovernmental meetings with high-level representatives from each participating country. "They go through the policy summary and approve it word by word," says Kevin. "Politics enter in very strongly at this stage. Scientists are there to defend the science and what can be said; the politicians get into the act about how it can be said."

During the intergovernmental meetings for the last IPCC cycle, Kuwait and Saudi Arabia registered strong disagreement with some sections of the WG1 report. According to Kevin, "U.N. rules are to try to arrive at a consensus rather than a majority vote." On the final night, "We were supposed to be done at 10:00 p.m., and we were still working at 3:00 a.m., long after the translators had gone home." Ultimately, the WG1 technical summary was "accepted" but not approved line by line, thus leaving the authors to make final changes without another review pass.

Climate news the U.S. can use

Ben Felzer.

As IPCC work continues, the U.S. National Assessment is putting the finishing touches on its report, which will be published this summer. At that point, Ben Felzer's job will be completed. Ben, who works for the U.S. Global Change Research Program through JOSS, has helped coordinate much of the report's climate sections and prepared dozens of graphics for the document. He explains that the assessment answers the questions, "What are the consequences of global warming for the U.S.? And how should we deal with these consequences?" Specifically, the assessment considers the effect of climate change on five economic sectors--agriculture, coastal areas and marine resources, forests, human health, and water resources--and 20 geographic regions. One of these regions is a crosscutting area that represents native peoples and native homelands.

Mandated by Congress in 1990, the report is being overseen by the Office of Science and Technology Policy and the National Assessment Synthesis Team, funded mainly through the EPA, NASA, NSF, and the Departments of Energy, Agriculture, Interior, and Commerce (NOAA). It is the first comprehensive climate assessment for the United States, although other countries, such as Canada and Great Britain, have done related types of assessments.

One of the U.S. report's strongest features, and a rarity for assessments, is that the process involved stakeholders in each region: farmers, resource managers, and others who will have to deal directly with climate change. They participated alongside scientists at regional meetings, and their input shaped the final product. For instance, stakeholders helped decide which of the economic sectors their region should focus on.

Kathy, Linda, and Jerry have each been involved in preparing this assessment as well as the IPCC report. Discussing a session she coordinated for a Central Great Plains regional meeting, Kathy says, "I thought it was really helpful to have the stakeholders involved and providing feedback about their information needs. Although the particulars of what could happen are important, they communicated that it's more important to them to understand the general nature of the risks, and the potential range of climatic variations they may face, than it is to have the details of any particular scenario. They're already thinking adaptation; that's because they deal with climate variability every year, every day." Linda participated in three of the regional meetings and appreciated "communicating with a broader range of people about climate scenarios."

The project navigates some tricky scientific territory, since regional effects of climate change are notoriously difficult to project. Scenarios, both climatic and socioeconomic, are important tools for the assessment. They indicate what could happen in the next century in several ways--for instance, identifying how societal vulnerabilities may change over time, and how they might be modified by adaptation choices. The climate-scenario approach involves:

  • projecting historical trends, such as the increased frequency of El Niño, into the future;

  • painting "what if?" scenarios for different increases in average temperature, changes in precipitation; and

  • presenting results from mid-1990s versions of the Canadian Global Coupled Model and the coupled model from England's Hadley Centre for Climate Prediction and Research.

    The Canadian and Hadley Centre models were chosen largely because their output, which included daily data, was available several years ago, giving the preparers enough time to translate it to the regional level. For this translation, the assessment relied heavily on the Vegetation/Ecosystem Modeling and Analysis Project, based largely in CGD (a separate story on VEMAP's work appears in this issue under Science Briefing).

    Ben and his colleagues don't expect this assessment to be the last word on climate change in the United States. "The idea is to get people thinking about these issues," says Ben. By this measure, the assessment is already a success. "The final report isn't going to say too much that the stakeholders don't know already," says Kathy. "I think they were more interested in being in the process. I think that's more important to them than any written report."

    Once more, with feeling?

    Although assessment work gives you prestige and makes you catch up on the literature, it doesn't get you any grants or produce any journal articles. "There's a high burnout rate," says Kevin. He notes that Ben Santer (Lawrence Livermore National Laboratory) and Neville Nichols (Australian Bureau of Meteorology) both opted out of IPCC 2000 after playing key roles in 1995. "It's very demanding. You just get finished with one [report] and you have to start the next."

    Tom Wigley.

    CGD senior scientist Tom Wigley had been heavily involved in IPCC from its beginning. He's sitting out the 2000 process, except for minor inputs to WG1 and a short contribution on uncertainty for the WG3 report. He wonders if the IPCC process is operating on too short a time scale. "Do we need to update the science comprehensively every four or five years? How different is the third assessment going to be from the second? If it's different in only a few areas, why don't we do some special reports in those areas?"

    A low-key lobbying effort is now under way among IPCC scientists to shift the schedule from every five to every seven or even ten years. "The policy makers seem to want new information nonstop," says Jerry. "Five years in the political arena is a long time, but the scientists are saying, 'It's too much, too fast.' "

    "It's a monumental effort," agrees Tom. "I don't know if anybody outside this community realizes what a massive job this is." Still, he adds, "I think it's more or less a duty for people who are authorities in these areas to participate at least once."

    Bob Henson

    On the Web:
    IPCC home page
    U.S. National Assessment

    A sneak preview

    Although the next IPCC report isn't ready until next spring, our in-house experts provide a taste of what's likely to appear. Kevin Trenberth believes there's no particular bombshell in the works from Working Group 1. "At the moment there's no banner that IPCC 2000 would come out with that would grab as much attention [as in 1995]."

    That's not to say there won't be findings of substance, Kevin adds. The past five years of global heat have bolstered the circumstantial case for greenhouse-induced warming, and the long-standing conflict between satellite and surface data has been largely resolved (see the February 2000 Staff Notes Monthly). "The observational elements are much more compelling, and the models have gotten much better. The statistical analysis tools for detection and attribution of climate change have also gotten stronger. So there have been developments across a broad front."

    Linda Mearns says that uncertainty and its expression will be a big theme of the 2000 report. "I think it'll be an area of rapid methodological development in the next few years. You can't know what's certain unless you know what's uncertain."

    According to Jerry Meehl, "The three things everyone wants to know about are the three things we can say the least about, due to model limitations: hurricanes, midlatitude storms, and ENSO [El Niño/Southern Oscillation]." The IPCC 2000 report will reflect some signs of progress, though. As noted in the February issue of Staff Notes Monthly, the latest hurricane models--which can now be embedded within global models--are hinting at greater tropical cyclone intensities for parts of the Pacific basin in a world of doubled carbon dioxide. And some models are beginning to show how El Niño and La Niña might intensify in a warmer world.

    Results of the U.S. assessment are embargoed for now. According to Ben Felzer, the report will include both benefits and costs of climate-change impacts. He summarizes the climate scenarios used to prepare the assessments: Temperatures will climb in general, but more quickly at high latitudes, at night, and in the winter. Rainfall and snowfall are tougher to call. "Global precipitation will rise, but we have no idea about regional trends," says Ben. The report does draw some reasonable inferences on how temperature and precipitation trends might interact, altering seasonal snowfall patterns, for instance.. •BH


    In this issue...
    Other issues of Staff Notes Monthly


    UCAR | NCAR | UOP

    Edited by Bob Henson, bhenson@ucar.edu
    Prepared for the Web by Jacque Marshall