In the final minutes of its plunge toward the moon, NASA's LCROSS spacecraft spotted the brief infrared flash of a rocket booster hitting the lunar surface just ahead of it – and it even saw heat from the crater formed by the impact. But scientists remain puzzled about why the event did not seem to generate a visible plume of debris, as was expected.
As hundreds of telescopes and observers watched, the highly publicised NASA mission to search for water on the moon reached its grand finale at 0431 PDT (1131 GMT) with a pair of high-speed crashes into a lunar crater named Cabeus.
During the crucial moments at NASA's Ames Research Center in Moffett Field, California, scientists and engineers with LCROSS (Lunar Crater Observation and Sensing Satellite) peered in silent concentration as successive images of the crater grew larger on their screens.
Nearby, some 500 bleary-eyed visitors that had gathered overnight outside mission control were watching the same pictures on a giant outdoor screen.
Yet, immediately after the scheduled impact time, there was no obvious sign of the spectacular explosion that many were expecting. "Impacting into the moon is an unpredictable business at best," Anthony Colaprete, principal investigator for LCROSS, said in a post-impact briefing.
Colaprete did not offer definitive word as to why the visual camera apparently did not detect the event but added there were interesting changes in spectroscopic data taken by the spacecraft that might have been produced by a debris cloud. "I'm not convinced that the ejecta is not in the data yet," he said.
A worst-case scenario would have occurred if the rocket hit bedrock rather than loose, gravelly soil. In that case, the debris plume might not have reached the minimum 1.5km altitude needed to catch the sunlight and be seen by LCROSS.
Because of the angle of the crater, the plume would have needed to rise to 2.5 to 3km in order to be seen by telescopes on Earth. A 10-km-high plume was expected.
The impact was monitored by the Hubble Space Telescope, which has not yet delivered its data. Several major observatories were also watching for signs of impact, including the Keck and Canada-France-Hawaii telescopes on Mauna Kea, neither of which saw a plume. One positive report came from Kitt Peak Observatory in Arizona, where a flash of visible light revealing the presence of sodium was recorded during the impact.
"I think we're all a little bit disappointed that we didn't see anything," David Morrison, director of NASA's Lunar Science Institute, told New Scientist.
Neither here nor there
Regardless of its ultimate scientific return, today's outcome will likely go down as one of the more bemusing episodes in NASA's long history of lunar missions. While the spacecraft appeared to be working as expected and in contact with mission controllers, it clearly did not deliver the views that scientists and spectators were hoping for.
Unlike a catastrophic failure, such as Mars Polar Lander in 1999, or a euphoric success, such as the spectacular 2005 collision of the Deep Impact mission with Comet Tempel 1, the non-detection seemed to leave officials unsure of how to react.
The big question that planetary scientists hope will be answered is: are there significant quantities of water ice on the moon? Last month, water was discovered in the lunar soil, but the amounts detected were relatively small.
A long standing mystery is whether dark craters such as CabeusMovie Camera could act as cold traps, capturing water molecules that are liberated when comets strike the moon. Data from the Lunar Prospector mission, which flew in the late 1990s, indicate high concentrations of hydrogen in Cabeus. The hydrogen could belong to water ice mixed in with the rock and soil in the crater's depths.
LCROSS was designed to look for the signature of water and other molecules as it flew into the debris plume of the rocket impact. It should also have executed a sideways turn one minute prior to its own impact to see the molecular constituents of the impact backlit by the sun.
Without a plume to study, scientists will have less of a handle on the question but Colaprete says the spectroscopic data may be enough to spy the constituents of water. "It will probably take two weeks to get a yes or no answer on water," said Michael Bicay, director of science at Ames.