Page images
PDF
EPUB

and implications stemming from the models and issues currently being debated in the scientific literature.

At one end of the spectrum are those statements of high certainty that do not stem directly from climate models but are largely based on measurements, observations, and well-established scientific principles.

For example, scientists agree that greenhouse gases, such as CO2, absorb and readmit infrared radiation; that atmospheric concentrations of CO2 and other greenhouse gases are significantly above preindustrial levels; and that increased concentrations of greenhouse gases exert a global warming influence.

The global climate models in turn incorporate both existing knowledge, as well as assumptions about inadequately understood matters, such as the role of clouds in global warming, to predict future climate.

The forum participants felt certain some of these predictions were more probable than others. For example, an increase in the global surface temperature from one to 3.5 degrees Fahrenheit by the year 2050, a reduction in northern hemispheric sea ice and a rise in global sea level were all considered to be very probable results of increased concentrations of greenhouse gases.

Scientists, on the other hand, have less confidence in the probability of other outcomes such as increased mid-latitude drought and changes in tropical storm intensity.

In short, Mr. Chairman, the models are considerably better now than they were a decade ago at predicting future climate change, and the scientists involved have varying degrees of confidence in the models' capability to predict certain outcomes.

The accuracy of their estimates, however, is limited by a number of factors, including inadequate understanding of climate processes and feedback mechanisms, limited computing power, and the necessity of using simplifying assumptions and structures to deal with these limitations.

These limitations do not change the likelihood that climate will respond to increased greenhouse gas emissions. They do however limit scientists' ability to predict with certainty how the climate will respond, how much warming will occur, how soon it will occur, and what the impacts will be on regional climates.

Several on-going efforts are underway to address the current limitations of these models. Scientists are working to develop models that better reflect interactions among various climate components, and that provide a more complete understanding of key processes, such as cloud feedback.

Approximately a third of the U.S. Global Change Research program's 1995 budget was devoted to conducting research aimed at improving scientific understanding and ability to model such proc

esses.

Steps have also been taken to provide increased computer capacity needed to improve the scientists' ability to run these complex models without having to make the simplifying assumptions that reduce their accuracy.

Given the complexity of these challenges, scientists believe that significant reductions in the current uncertainties of predicting cli

*

mate change will require sustained efforts over the next decade or longer.

Mr. Chairman, this concludes my remarks. I would be pleased to answer any questions.

[The prepared statement of Mr. Guerrero follows:]

PREPARED STATEMENT OF PETER F. GUERRERO, DIRECTOR, ENVIRONMENTAL PROTECTION ISSUES, RESOURCES, COMMUNITY, AND ECONOMIC DEVELOPMENT DIVISION, UNITED STATES GENERAL ACCOUNTING OFFICE

Mr. Chairman and Members of the Subcommittee:

We appreciate the opportunity to appear before the Subcommittee as you focus on issues concerning the use of climate change models. Over the past century, human activity has increasingly emitted heat-trapping greenhouse gases, such as carbon dioxide, into our earth's atmosphere. As these emissions have grown, so too has concern over how they will affect our climate in years to come. Temperature increases could begin to melt the polar ice caps, raise the sea level, and alter patterns of precipitation. These effects could lead to serious consequences such as severe flooding. On the other hand, actions to reduce greenhouse gas emissions in order to avoid potential catastrophic effects of global warming could have significant economic consequences, and must be based on a sound understanding of the issues. One such major issue that should be considered when discussing the implications of global warming and the appropriate actions to control greenhouse gas emissions, is the range of projected temperature increases and the degree of uncertainty in these estimates. In July of this year, we reported on the limitations of general circulation computer models used to make such predictions.1 These sophisticated computer models consist of complex mathematical equations that represent various climatic processes and interrelationships among variables, such as seasonal changes in sunlight and global air currents. Our testimony today is based on our July 1995 report.

In general, we found that general circulation models are better now than they were a decade ago at predicting future climate changes. Nevertheless, the accuracy of the models' estimates is still limited. Specifically, we reported that:

-For general circulation models, as for other computer models, the quality of the output depends upon the quality of the input-the models are only as good as the data and scientists' understanding of how the climate system works. One limiting factor is that the models currently provide only incomplete or inaccurate representation of some of the processes, such as cloud formation, affecting climate because scientists do not fully understand how the climate system responds to these processes. As a result, models can produce different projections of future climatic conditions, such as global temperature.

-A second major limiting factor is insufficient computing power to process the vast quantities of data required to more accurately simulate changes in the global climate. Modelers try to overcome this limitation by introducing assumptions that deliberately simplify some operations in order to free computer time for other, more critical operations. These simplifications can affect the accuracy of the models' estimates.

-To improve the accuracy of general circulation models' estimates, scientists are developing models that better incorporate the processes affecting the climate system and that better reflect the interactions between or among the ocean and the atmosphere. They are also developing larger and faster computers that could process data for smaller areas. Given the complexity of the climate processes that need to be incorporated in the models, scientists believe that significant reductions of the uncertainties in projecting changes and trends in the climate will require sustained efforts that are very likely to require a decade or more.

BACKGROUND

General circulation models are the most highly developed tools available to help understand the global climate system's response to greenhouse gas emissions. Scientists use three types of general circulation models to predict climatic change: atmospheric, oceanic, and coupled. In general, atmospheric models predict the physical behavior of the atmosphere. Oceanic models represent the physics of the ocean. Coupled models, which scientists regard as the most advanced type of model, physically join the atmospheric and oceanic models. In the United States, the development of

1Global Warming: Limitations of General Circulation Models and Costs of Modeling Efforts (GAO/RCED-95-164, Jul. 13, 1995).

general circulation models is supported through the coordinated efforts of the U.S. Global Change Research Program and five federal agencies.2

During the Forum on Global Climate Change Modeling, held in October 1994, scientists agreed that the buildup of greenhouse gases is creating an enhanced greenhouse effect that will lead to global warming.3 They estimated that the surface temperature of the earth will rise by 1 to 3.5 degrees Fahrenheit from 1990 to 2050 if emissions continue to grow without restriction. More recently, the International Panel on Climate Change reported that human activities are increasing the atmospheric concentration of greenhouse gases and that this increase will raise the global mean temperature between 1.8 and 6.3 degrees Fahrenheit by 2100.4 FACTORS LIMITING THE ACCURACY OF MODELS' ESTIMATES

General circulation models' estimates of future climatic changes are considerably better than they were a decade ago. The models have demonstrated skill in simulating many aspects of the observed climate, providing useful indications of future climatic conditions. For example, atmospheric models have demonstrated some skill in portraying aspects of atmospheric variability, such as the surface temperature of the sea. Oceanic models have simulated the general circulation of the ocean, including the patterns of principal currents. Coupled models, though still prone to small scale errors, have simulated the current climate on a large scale as well as portrayed atmospheric and oceanic behavior over large regions.

Though much progress has been made, the models remain limited in their ability to estimate, with desired accuracy, the magnitude, timing, and regional distribution of future climatic changes. These limitations stem from scientists' imperfect understanding of the climate system and computers' insufficient capacity to perform the detailed calculations needed to make more precise estimates.

Incomplete or Inadequate Representation of Processes Affecting Climate

According to the U.S. Global Change Research Program, general circulation models include the most important processes-such as radiation, convection, and water vapor-that affect the climate. However, the extent to which they incorporate and accurately represent these processes and their interactions varies and can affect the accuracy of the models' estimates. For example, some climate processes are not included or fully incorporated, even in the more advanced coupled models. And for some of the more important climate processes and interactions that are included in models, their representation is less than adequate primarily because scientists do not fully understand the climate system.

Atmospheric and oceanic models include fewer processes than coupled models where the oceans and the atmosphere meet, such as oceanic pressures, climatic fluctuations, and the effects of winds at the surface of the oceans. Consequently, these models' simulations are more limited and, in some cases, less accurate. For example, a 1991 test of atmospheric models produced systematic errors in the projections of sea level pressure, temperature, wind, and precipitation.

Coupled models more accurately simulate current climatic conditions than either atmospheric or oceanic models, but their estimates of temperatures and precipitation still deviate from actual conditions. For example, in an experiment conducted by the National Center for Atmospheric Research, the models estimated wintertime ocean temperatures were 7 degrees warmer than observed temperatures for the icebound region of Antarctica and 9 degrees colder than observed temperatures for the tropics. Scientists believe that the deviations stem from gaps in their understanding of the interactions between atmospheric and oceanic variables.

Although today's general circulation models include many of the most important feedback mechanisms (e.g., water vapor and clouds) they do not yet adequately represent the interactions of these mechanisms with greenhouse gases. Such interactions can amplify, dampen, or stabilize the warming produced by increased concentrations of greenhouse gases. Modelers do not fully understand the effects of feedback mechanisms and have not learned how to represent them with sufficient accuracy. Modelers have clarified the role of water vapor and improved their ability to model its effects. However, they are still seeking to understand and accurately

2 These five agencies are the Department of Energy, the National Aeronautics and Space Administration, the National Science Foundation, the National Oceanic and Atmospheric Administration, and the Environmental Protection Agency.

The purpose of the Forum, convened to respond to GAO and White House Office of Science and Technology Policy questions, was to produce a consensus document on issues concerning the use of climate models and to provide policymakers with information on future climatic changes. 4 The International Panel on Climate Change was established in 1988 by the United Nations Environment Programme and the World Meteorological Organization to assess scientific and technical information about climatic change.

[merged small][ocr errors]

model the effects of clouds, which have the greatest potential of all feedback mechanisms to amplify or moderate global warming.

Insufficient Computing Power

The accuracy of general circulation model estimates are further limited by computers' insufficient capacity to store and analyze the vast quantity of data required to accurately simulate global climate changes on a regional scale. In an effort to overcome these limitations, modelers introduce assumptions into their models that deliberately simplify some operations in order to free the models' capacity and time for other, more critical operations.

One such simplification used in certain models, referred to as the cold start error, assumes that the oceans did not absorb greenhouse gas emissions before 1985. As a result, these models do not adequately account for the oceans' reduced capacity to absorb these emissions in the future. In fact, the oceans will reach their capacity possibly decades sooner than the models have predicted. They will then deflect more of the heat-trapping emissions to the atmosphere, thereby enhancing global warming more rapidly than the models predict. Scientists estimate that cold start error causes models to underestimate temperature changes, resulting in projections of average annual temperatures that differ by as much as 0.7 degrees after 50 years. Months of additional computer time would be required to address just this one existing limitation.

Another simplification that modelers use to deal with insufficient computer capacity is to divide the earth into large segments, or grids, for analytical purposes. Although the grids, which cover an area about the size of South Carolina, enable the models to depict large-scale regional effects in relatively large, homogeneous regions, they do not allow modelers to incorporate detailed regional features to forecast climatic changes for smaller, less homogeneous regions.

Improving Models' Estimates

A number of efforts are under way to improve the accuracy of general circulation models' estimates. Scientists are working to develop models that incorporate more of the processes affecting the climate system-particularly cloud formation processes-and to better reflect interactions among various components of the climate system, including interactions between or among the ocean and the atmosphere. Scientists are also developing larger and faster computers to manipulate data for longer periods of time and to better understand regional effects resulting from using smaller grids. In addition, they are collecting more data and conducting more research on the processes affecting climate and improving the international exchange of such data. Various international programs, such as the World Climate Research Programme and the Global Climate Observing System, also currently have research programs under way to improve the accuracy of the models' estimates. CONCLUSIONS

Although the accuracy of general circulation models has improved over the past decade, these models are still limited by incomplete and inaccurate representations of the processes affecting climate and by insufficient computing power. While these limitations do not change the likelihood that the climate will change in response to increased greenhouse gas emissions, they do limit the ability to predict with certainty how the climate will respond-how much warming will occur, how soon it will happen, and what regional impacts will occur. Efforts are under way to collect and analyze data, improve representations of various climatic processes, and develop and use more powerful computers. These efforts, which are anticipated to take a decade or more, should improve the accuracy of the models.

Mr. Chairman, this concludes my prepared statement. We will be pleased to respond to any questions you or other Members of the Subcommittee may have.

Mr. ROHRABACHER. Dr. Mahlman.

5 This program was established as a joint undertaking of the International Council of Scientific Unions and the World Meteorological Organization, to foster an improved understanding of the climates variability and predictability.

This international organization was established in 1992 to monitor climatic changes and ob

STATEMENT OF DR. JERRY MAHLMAN, DIRECTOR, GEOPHYSICAL FLUID DYNAMICS LABORATORY, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION

Dr. MAHLMAN. Mr. Chairman, I am the Director of the Geophysical Fluid Dynamics Laboratory of NOAA. For over 30 years, our laboratory has been a world leader in modeling the earth's climate.

Today, I will evaluate scientific projections of climate change, as well as their current uncertainties.

We have long known that buildups of atmospheric carbon dioxide and other gases have the potential to warm earth's climate through the so-called greenhouse effect.

I will emphasize only the estimated climatic effects of these greenhouse gases for a time around the year 2050.

Because I speak with credentials as a physical scientist, I do not offer personal opinions on what society should do about these predicted climate changes. Societal actions in response to greenhouse warming involve value judgments that are beyond the realm of climate science.

Indeed, I would encourage your skepticism whenever you hear a climate scientist's prediction that is accompanied by a policy opinion.

At the onset, please recognize that a major international effort to assess climate change was completed in 1990. This is the IPCC Assessment.

Also, it has been recently updated in this 1994 supplement.

These are the most widely accepted statements ever on climate change.

A new IPCC statement will be completed in the next few months. I expect only small changes in its major conclusions, mainly concerning some increases in scientific confidence.

I strongly recommend your use of these scientific assessments as the foundation for your own evaluations. I also recommend their use as a point of departure for evaluating the credibility of opinions that disagree with them.

My information is derived from the strengths and weaknesses of climate models, climate theory, and widespread observations of the climate system.

Climate models have improved in their ability to simulate the climate and its natural variability. Unfortunately, important uncertainties remain due to deficiencies in our scientific understanding, as well as in our computer power.

However, I expect significant progress over the next decade.

However, let me say at the outset none of the uncertainties I will discuss here can make current concerns about greenhouse warming go away. This problem is very real and will be with us for a very long time.

I give my evaluation of current model predictions of climate change in the middle of the next century by using very simple betting odds. When I say virtually certain, I mean that there is no plausible alternative. In effect, the bet is off the books.

Very probable means I estimate a nine out of ten chance that this will happen within the range predicted.

Probable implies about a two out of three chance.

« PreviousContinue »