washington

Hope is fading that US ground controllers will be able to regain control of the $71-million Lewis demonstration satellite before it re-enters the Earth's atmosphere around mid-September.

The satellite, which is testing the feasibility of hyperspectral imagery of Earth as well as advances in spacecraft design, went into a spin on 26 August, shortly after launch, and radio contact was lost (see Nature 389, 10; 1997). Project managers at the California-based company TRW, Inc., which built and operated Lewis under contract to NASA, had hoped that the tumbling satellite's solar panels might absorb enough sunlight to recharge onboard batteries and bring the craft back to life. But as Nature went to press, this had not happened. NASA officials describe the chances of recovering the satellite as slim.

The high-profile project was also an experiment in mission management, with the private contractor TRW operating the craft with minimal involvement from NASA. To save money, TRW employed a lean management team and minimal tracking facilities, and some observers argue that this may have contributed to the problem.

A NASA source alleges that the Lewis control room in Chantilly, Virginia, was left unattended for several hours during the period that the satellite began to spin. Officials at NASA and TRW declined to confirm or deny this. One NASA official says the accident resulted from “a series of very low probability events happening at the wrong time”. The official also argues that the loss cannot be blamed on either the private contractor or NASA's recent policy of “better, cheaper, faster” missions. “What went wrong on this project could have gone wrong on any project,” he said. The Lewis mission has been strongly supported by Dan Goldin, the NASA administrator, who is keen to see greater private sector involvement in space activities.

Although TRW is managing the recovery effort, NASA has brought its own expensive resources to bear. The large antennas of the Deep Space Network and orbiting Tracking and Data Relay Satellites were enlisted last week to try to force the spacecraft to point its solar arrays at the Sun. But because the spacecraft is spinning, its radio receiver is not pointed in any one direction, making it difficult to send commands.

The loss of the Lewis spacecraft would be a blow to Earth scientists, who have been eagerly awaiting test results from two advanced hyperspectral imagers; these have been flown on aircraft but never on a civilian satellite. Hyperspectral imagery is considered to be the next major advance in satellite remote sensing, because it allows imaging using hundreds of spectral bands, whereas conventional Earth observation satellites use only a dozen or so bands.

The Lewis craft was intended to test whether hyperspectral imagery would be able to distinguish between different types of trees in a forest, and to assess their health on the basis of leaf chemistry. The imagery technique also has potential applications in mineral mapping, agriculture and planetary exploration; indeed, one of Lewis's two imagers was originally developed to fly on a NASA Pluto mission, which has since been put on hold.