Senior U.S. government officials have narrowed their list of options for containing the cost of a new generation of civil-military weather satellites, but meanwhile are telling Capitol Hill that the program now faces a review that in theory could lead to its termination.

In September, the U.S. Air Force notified Congress that the cost of the National Polar-orbiting Environmental Satellite System (NPOESS) has risen by 15 percent. Such notification is required under a U.S. military procurement law known as the Nunn-McCurdy provision.

Now service officials are warning congressional staffers that the NPOESS cost-growth projections have reached 25 percent, which under Nunn-McCurdy triggers a recertification review, according to sources who track the program. Such reviews must justify a program’s continuation based on its importance to national security, the lack of viable alternatives, and evidence that the problems that led to the cost growth have been wrestled under control.

Lawmakers are expected to receive formal notification of the latest NPOESS Nunn-McCurdy breach in January, the sources said. The Pentagon then has until May to conduct the recertification review and report back to Capitol Hill, the sources said.

The Air Force and the U.S. National Oceanic and Atmospheric Administration (NOAA) are jointly funding the NPOESS program, which will replace separate lines of polar-orbiting weather satellites that traditionally have served military and civil users. Northrop Grumman Space Technology of Redondo Beach, Calif., is the prime contractor.

The NPOESS program previously was expected to cost $7.4 billion, including six satellites, sensors and ground systems. Launches were to begin in 2009. But NOAA Administrator Conrad Lautenbacher and Air Force Undersecretary Ronald Sega told the House Science Committee Nov. 16 that they now expect the first NPOESS satellite to launch no earlier than 2012 due to technical difficulty with the system’s sensors. They also said the program’s cost likely would rise by $2 billion to $3 billion.

Sources said the cost-growth figure is now expected to surpass $3 billion.

Lautenbacher and Sega, as well as NASA Administrator Mike Griffin, met Nov. 22 to discuss options for mitigating the NPOESS cost growth, the sources said. NASA is a junior partner in the NPOESS program, with responsibility for technology insertion. The group has asked the Pentagon’s Cost Analysis Improvement Group to develop cost estimates for its short list of alternatives, the sources said.

One option is continuing on the schedule outlined by Lautenbacher during the hearing, the sources said.

Another entails dropping an instrument called the Conical Scanning Microwave Imager/Sounder, which is being developed by Boeing Satellite Systems of El Segundo, Calif. That sensor is still early in its development, but its projected size has increased significantly, sources said. Accommodating the sensor will require a larger NPOESS satellite platform than planned, and this has become a factor in the projected cost growth, the sources said.

The Conical Scanning Microwave Imager/Sounder is designed to monitor atmospheric temperature and humidity, cloud cover and ocean-surface wind speeds. The sources said the wind-speed measurements could be replaced by an operational version of the experimental Coriolis satellite, which was launched in 2003 and featured an instrument called Windsat built by the U.S. Naval Research Laboratory in Washington.

Joe Tedino, a Boeing spokesman, said in a written statement that the company has made significant technical progress on the Conical Scanning Microwave Imager/Sounder over the course of the past year.

“We believe we are in good shape on the program,” Tedino said.

Boeing will work with Northrop Grumman to “meet program and funding requirements over the next few years,” Tedino said.

The Cost Analysis Improvement Group also has been asked to assess the cost of buying four rather than six NPOESS satellites, and to determine how many satellites could be purchased for the previous $7.4 billion price tag, the sources said. Without more funds, it is likely that only two satellites could be purchased, the sources said.

The technical difficulty with the sensors and the delay to the first NPOESS launch has led program officials to shuffle the schedules of other satellites to avert a potential gap in weather coverage. The NPOESS Preparatory Project, a NASA-led mission that originally was scheduled to launch in 2006 to test key NPOESS instruments and support global climate research, likely will be delayed about two years, Lautenbacher said at the hearing. The final launch of NOAA’s current-generation polar-orbit satellites likely will be delayed as well, he said.

Meanwhile, the instrument payload aboard the NPOESS Preparatory Project satellite could change, the sources said. Currently that payload includes the Visible Infrared Radiometer Suite, which is built by a unit of Raytheon Co. in Santa Barbara, Calif., and which has encountered technical problems that have been cited as a major factor in the NPOESS cost growth.

If program officials run into continued schedule delays with that sensor — designed to provide imagery and take ocean-surface temperature and color readings — it could be dropped from the NPOESS Preparatory Project and launched about a year later on a separate satellite, one source said. Building a second satellite platform for the first Visible Infrared Radiometer Suite likely would cost about $100 million, the source said.

John Leslie, a spokesman for NOAA, and Sally Koris, a spokeswoman for Northrop Grumman, said in written statements that the government and industry team plan to work together closely to “re-plan the NPOESS program and ensure that it will deliver this much-needed capability to military and civilian users.”