Chairman WALKER. Thank you very much. We appreciate your testimony. We will now move to Mr. Brad Hathaway from the U.S. General Accounting Office. Mr. Hathaway?

Mr. Hathaway?

STATEMENT OF MR. BRAD HATHAWAY, U.S. GENERAL
ACCOUNTING OFFICE, WASHINGTON, DC.

Mr. HATHAWAY. Mr. Chairman, it's an honor to be here today to discuss the previous work that we at GAO have done on the lifecycle cost of EOS, and our ongoing work on the research community to analyze EOS data.

We reported, and you mentioned in your opening statement, Mr. Chairman last June, that the life cycle of the baseline EOS program at that time would be about $33 billion.

We'll report further to you on the cost reduction efforts after the release of the Fiscal Year '97 budget when we get a chance to review the data and the submission from that.

The preliminary results of our ongoing work into the research community for analyzing EOS data is that we believe that NASA's current EOS basic research community is relatively small compared to pre-EOS missions, and it's uncertain whether NASA can successfully expand it within future budget constraints.

If NASA is not successful, there might be an imbalance between investigations and the EOS data.

Let me discuss costs a little bit first.

Initial planning for EOS, as you know, occurred during the 1980s, when NASA's budget_was increasing each year. In 1991, NASA's plan for spending on EOS was $17 billion through the year 2000.

In 1992, we reported on a growing gap between NASA's program plans and its probable future budgets. Congressional concerns about EOS's affordability caused NASA to change EOS's focus to global change on the earth's climate.

This estimated cost through the year 2000 was reduced to $11 billion and then to $8 billion and then to $7.25 billion.

Over the last several years, NASA was successful in closing the affordability gap that we had identified. However, by the President's 1996 budget submission, that gap had reopened. Although much of the budget debate focused on that first decade of the program, EOS is planned to continue for another 20 years beyond that.

In June, as I said, we reported life cycle cost estimates for the then-EOS baseline program, totalling about $33 billion from Fiscal Year '91 through Fiscal Year 2022.

NASA recognized that the program, as designed in early 1995, was not affordable. It was already studying ways to substantially reduce costs. They continued that effort, and I believe you'll be hearing more about that later on this morning.

Let me turn to NASA's strategy to build an EOS research community.

NASA originally solicited proposals for EOS interdisciplinary science investigations in 1988 and selected 29 teams.

In our ongoing work, we found that although NASA has not determined what size research community is sufficient, last year it

recognized that more basic research teams were needed. NASA solicited proposals in September, to address specific interdisciplinary science issues that are not well covered by existing investigations. However, the announcement largely precluded investigators from analyzing data from EOS missions. Instead, NASA asked for them to propose research that primarily uses existing data sets. NASA intends to support this research by freezing the budgets of some of EOS's interdisciplinary science investigations.

EOS's 29 teams needs to be put in context.

Over the years, NASA has stated that about 10,000 earth scientists might use EOS data. However, we found that for this type of work, scientists generally analyze data when they are paid to analyze data.

We reviewed 172 journal articles about two pre-EOS earth science missions-the Upper Atmosphere Research Satellite, UARS, and the TOPEX/Poseidon. Publicly-funded investigators wrote all but ten of those articles.

Our observation that the size of the EOS research community is relatively small is based on a comparison with the number of science teams associated with those two missions and the ratio of the number of science teams to the raw data acquisition rates.

UARS consists of ten instruments that are measuring the composition and temperature of the upper atmosphere, atmospheric winds and energy from the sun.

In the EOS era, solar energy and atmospheric chemistry measurements will be made principally by three instruments and the CHEM spacecraft. Currently, only 12 instrument and interdisciplinary science teams are associated with these instruments and the CHEM spacecraft.

In contrast, UARS research is conducted by 62 teams.

TOPEX studies the circulation of the world's oceans. NASA and its French partner selected 38 science teams.

In the EOS era, the follow-on mission to TOPEX is Radar-ALT, and although an instrument team hasn't yet been selected, only 7 of the 29 inter-disciplinary science teams currently plan to use Radar-ALT data in their investigations.

Now there's a striking difference between the number of science teams and the volume of data for pre-EOS missions and EOS.

Taken together, the number, but not necessarily the size, of the UARS and TOPEX science teams is a little larger than all of the EOS teams, while EOS's data rate is close to a thousand times greater than the combined data rate of UARS and TOPEX.

Now although we recognize that the data rate comparison is at best a rough gauge, the difference is too large to ignore.

Achieving the best science return on the large investment in EOS will require a proper balancing of space systems, information systems and researchers. It's not clear to us that NASA will achieve this balance because the science community, funded to do basic research under EOS, is relatively small, and although NASA recognizes the need to increase the number of basic researchers, we remain concerned about NASA's ability to fund more basic research through future savings that may not materialize or that may be absorbed by budget reductions.

That completes my statement, Mr. Chairman.

[The prepared statement of Mr. Hathaway follows:]