Program Descriptions

Descriptions of the focused USGCRP research programs for which the primary emphasis is global climate change modeling are attached; note that for several of these programs, global climate modeling is only one aspect of the program's efforts.2

The attached table provides a summary of the FY-92 to FY-94 expenditures for global climate change modeling activities within their USGCRP programs. In addition, a few relevant climate change modeling activities are included in the totals by drawing from the broader set of agency activities conducted in support of environmental and general scientific research. As indicated in the notes to the table, the totals listed focus closely on global modeling activities relating to long-term climate change and do not include the extensive efforts involved in observations, data management, and process studies that are a necessary foundation for the testing and application of the models.

2 Note added for November 14 submission of report for Congressional hearing: these one page descriptions are not included with this memo in that the focus is more on the overall program.

Agency

FY-92
(M$)

Table: U. S. Government Agency Expenditures' for Global Climate Change Modeling for FY-92 to FY-9423

1

Note that these totals include expenditures for only DOC/NOAA, DOE, EPA, NASA, and NSF; expenditures by USDA and USGS that use results from
global models are generally small and are not included. The DoD also carries on global modeling activities, but they are focused on simulation of the weather
and seasonal to interannual variations (as are additional activities at NOAA, NASA, and NSF). The Electric Power Research Institute and other non-
government organizations have also provided funding for climate change research, mainly in the form of computer resources to researchers in the U.S. and
overseas. Columns may not total due to rounding.

2 The expenditures here include those for projects whose primary focus is modeling the global atmosphere, oceans, and/or land surface for the intent of
simulating changes in the global climate or changes in the chemical cycles and budgets of species forcing global climate change (e.g., carbon dioxide, etc.).
The totals include costs for testing, verification, application, and interpretation of the results from global models. Costs for computer resources are not fully
included because some agencies centrally fund computer resources for multiple purposes and then allocate time through a process separate from the budget
process. The totals listed exclude costs relating to developing parameterizations of particular processes to be included in the models (considered to be within the
understanding components of the USGCRP budget), costs relating to assembling and interpreting observations (considered to be data assimilation in support
of multiple USGCRP objectives), costs relating to simulation of the weather and seasonal to interannual fluctuations of the climate, costs for study of
paleoclimates, and costs related to improving numerical solution methods.

'For comparison, the United Kingdom Department of the Environment provides about 13M pounds per year (about $20M/yr) to support the climate change effort of the Hadley Centre (which is one of the top few foreign modeling centers).

4 NASA support for modeling activities that focus on assimilation of data amounted to $8.2M, $10.3M, and $10.8M in FY-92, 93, and 94. Funding by DOE and NSF for such activities were generally less than $0.5M.

On behalf of the Subcommittee on Global Change Research (SGCR), I would like to thank you and the General Accounting Office for providing the opportunity for the five agencies supporting research on global climate system modeling to review the draft version of the new GAO report entitled "GLOBAL WARMING: Limitations of General Circulation Models and Costs of Modeling Efforts." The SGCR is an interagency committee that has responsibility for oversight of the U. S. Global Change Research Program (USGCRP), for which most of the global modeling activities are undertaken. We welcome the interest of Congress and the GAO in reviewing the scientific limitations of the models and the progress being make on research to improve them.

Pursuant to your request, copies of the draft report were submitted for comment to representatives of the Department of Energy, the National Aeronautics and Space Administration, the National Science Foundation, the National Atmospheric and Oceanic Administration, the Environmental Protection Agency, and the White House Office of Science and Technology Policy. Review of the report was coordinated by Dr. Michael MacCracken, director of the Office of the USGCRP, to whom the letter was sent; that office supports the interagency SGCR.

In reviewing the draft version of the report provided to us on May 3, the five agencies have prepared a single, integrated response to reflect their comments, which were all similar. Briefly, with respect to the overall content of the report, the agencies believe that it is important for the GAO to provide additional perspective on the successes and accomplishments achieved with models in order to provide context for the description of model limitations that is enumerated in the GAO report. The state of modeling has advanced significantly in the past five years, and while models are not yet fully adequate to accurately address all of the questions being asked, they are quite successful in a number of ways. We believe that readers of the GAO report would benefit from having both limitations and successes described, as is done in international and national assessments. This could be readily accomplished by GAO by drawing from and then including in full our general comments and the report of the USGCRP Model Forum that was convened last October and which the GAO representatives attended. We include this Model Forum report as an

Mr. Peter Guerrero

Page 2

As a second general point, we think it would be helpful for the GAO report to indicate that the agencies have a coordinated program, with each agency contributing to the overall effort. This is apparent in the complementary descriptions of the agency programs and areas of emphasis in the appendices, but is not clearly articulated in the body of the report.

As is most effective in scientific research, there are indeed multiple groups pursuing similar problems, but normally in somewhat different ways so as to check each other's findings. Overall, we believe the area is underfunded as compared to international efforts, and for FY-1995 and 1996 have sought additional funding for these activities.

We appreciated the opportunity to discuss these comments with GAO representatives at the meeting on May 12. This letter represents the official submission of comments by the five modeling agencies included in the report.

Combined DOC/NOAA, DOE, EPA, NASA and NSF Comments on

the April 1995 Draft GAO Report on

Factors Limiting the Credibility of GCMs

Submitted May 22, 1995

Summary

The federal agencies involved in modeling the global climate system, and more broadly the government agencies participating in the U. S. Global Change Research Program, welcome the interest of Congress and the General Accounting Office in improving understanding of the research efforts underway to provide a continually improving knowledge base of the effects of natural factors and human influences on climate, the environment, and society. The GAO report has the potential of contributing to the effort of communicating the many challenges faced in making models that realistically represent the climate system. For this report to fully meet this potential, it is the joint position of DOC/NOAA, DOE, EPA, NASA, and NSF that it would be helpful to provide some perspective on what we have learned and can do as context for appreciating the limitations that still exist. We also think it important to point out the efforts that are underway to address the limitations that are indicated and to improve the overall performance of the model simulations.

Background on Modeling

The people of this nation and the world face an important challenge as their numbers grow and their demand on and use of resources increase in ensuring that they do not impose an unsustainable debt on future generations that is reflected in unprecedented environmental change that overstresses ecological and economic systems. Under the leadership of President Bush in 1989 and Congress in 1990, the U. S. Global Change Research Program was established with the goal of providing an improved predictive understanding of the Earth system. To achieve this capability, the USGCRP supports substantial efforts to document past changes, to observe present conditions, to assemble information on emissions and alterations form human activities, and to understand how nature's many processes control the behavior of the atmosphere, oceans, terrestrial and marine ecosystems, and the polar snow and ice. These activities are carried out in cooperation with nations around the world and provide the information base needed to underpin the efforts to make predictions from seasons to many decades into the future. Given the complexity and momentum of human activities and environmental processes, the effort to make predictions is truly one of the most challenging intellectual undertakings facing us. While progress can be expected to be slow, the choice of not seeking to understand the full implications of present and future human development activities would doom us to continual surprises for which the costs of responding would likely be large. It is in this light that we undertake efforts to predict the future climate.

While we can learn much about the behavior of the Earth system by examining how it has behaved in the recent and geological past, there has been no time in the past (except for the catastrophic conditions following major asteroid impacts) when the change in atmospheric composition has occurred as rapidly as it is being affected by human activities today. For this reason, and others, the past can provide insight about what factors will change the climate and by roughly how much, but cannot be used to project conditions into the future. For that, our only tool is to combine as much of our current understanding as we can into comprehensive numerical models of the Earth system and then to use these models (referred to as GCMs) to conduct "experiments" on what can be expected to happen in the event of, for example, tropical deforestation or unrestrained use of coal, oil, and natural gas, or widespread changes in land use through agricultural expansion or reforestation.