The U.S. national missile defense system deployed in Alaska and Hawaii is still not ready to be declared operational, according to the Pentagon’s top weapon tester.

While some testing success has demonstrated the feasibility of intercepting ballistic missiles launched at the United States, the system has yet to repeat those results consistently enough to inspire confidence in the Ground Based Midcourse Defense System, said David Duma, the Pentagon’s acting director of operational test and evaluation.

“Right now, that confidence is lacking,” Duma told the House Armed Services strategic forces subcommittee on March 15.

Ground testing indicates that the interceptors have the potential to protect against a “limited attack under certain conditions,” but recent difficulties in the flight testing program prevent the military from confirming those results, Duma said.

The Pentagon had planned to declare the Ground Based Midcourse Defense System operational in late 2004, but has held up that announcement as it conducts a series of command and control tests.

The Missile Defense Agency (MDA) and prime contractor Boeing Co. also have run into setbacks during recent flight tests when the interceptor rocket failed to take off. The rocket stayed on the ground in a December test due to a software glitch, and did not lift off in a February test due to a stabilizing arm that did not retract.

“From an operational mission perspective, these tests are failures,” Duma said.

However, in a real attack, operators could have recognized the rocket problems and fired another interceptor at the incoming ICBM, Duma said. This function has been simulated during testing on the ground, he said.

Duma was joined by Air Force Lt. Gen. Henry (Trey) Obering, the MDA’s director, and Lt. Gen. Larry Dodgen, commander of Army Space and Missile Defense Command, in testifying before the subcommittee.

Democrats on the subcommittee asked the toughest questions, but indicated that their concerns lay more with the pace of the missile defense deployment, rather than opposition to the overall concept of the system.

Rep. Silvestre Reyes (D-Texas), the ranking Democrat on the subcommittee, said that he was not discouraged by the recent test failures, but noted that the failures in the past three flight tests going back to 2002 would equal a strike out in baseball.

“I say regroup in the dugout and go get another turn in the batter’s box,” Reyes said. “But on the other hand, we should not pretend that [this] is an all-star system when it is still in development in the minor leagues. You can ruin a ball player by rushing him to the big leagues, and you can ruin this system by making it run before it can even prove it can walk.”

Rep. John Spratt (D-S.C.), who joined forces in 1999 with Rep. Curt Weldon (R-Pa.) to write legislation declaring that it is U.S. policy to deploy a national missile defense system, said he is concerned that testing thus far has not included data from sensors like the Cobra Dane radar in Shemya, Alaska.

Obering said geography has prevented the use of Cobra Dane during testing, but that it could be used in a real attack. The Pentagon does not have launch facilities that would allow it to fire targets past the Cobra Dane radar, but plans to do so later this decade after it develops and tests the ability to fire ballistic missiles from an aircraft, an MDA official said after the hearing.

MDA is satisfied that computer simulations can take the place of the Cobra Dane in tests , the official said.

Spratt also expressed concern that flight tests where the rockets actually lifted off have all featured a surrogate kill vehicle atop the interceptor. These tests may not have accurately demonstrated the stress of flight on the kill vehicle that will be used in real intercepts, he said.

Obering said the hardware and software on surrogate kill vehicle s are quite similar to that of the operational kill vehicle, which is expected to be used in the next flight test. The surrogate is a prototype version of the kill vehicle, which is built by Raytheon.