March 6, 2005


Twin Mars rovers in instrument mix-up (Jenny Hogan, 04 March 2005, New Scientist)

NASA's Mars rovers Opportunity and Spirit are identical twins - so alike that they even fooled NASA. Researchers have discovered that they sent the robots to Mars with an instrument meant for Opportunity inside Spirit and vice versa. [...] But something was worrying Ralf Gellert of the Max Planck Institute for Chemistry in Mainz, Germany. Gellert runs an instrument on the rovers called the alpha-particle X-ray spectrometer (APXS), which analyses the chemical composition of rocks. Opportunity had found higher concentrations of certain elements in the soil at its Meridiani Planum landing site than Spirit had at the Gusev Crater, but on a windswept Mars the concentrations should have evened out all over the planet. [...] "Now the reason for this is obvious," Gellert told New Scientist. "We found proof that we unintentionally swapped both instruments." Spirit is carrying the spectrometer destined for Opportunity, while Opportunity is in possession of Spirit's.

Although their designs are identical, each instrument is unique because of quirks in the materials they are made from. So before the rovers were launched, each instrument was calibrated using known rock samples. The measurements from each rover are then processed using the calibration files, but because of the mix-up, researchers were using the wrong ones. As a result, small errors have crept into the APXS results, affecting measurements of sodium, magnesium and aluminium abundance.

The corrections are very small because the instruments are so similar, he says. "None of our substantive scientific conclusions are affected." Gellert is relieved: "This turns out to be a lucky case where the data makes more sense now."

Squyres is "not embarrassed at all" about the slip-up with the rovers. "It was an easy mistake to make," he says. "It happened during some very busy and stressful times." He also says it is not fair to compare it to past mishaps because the spacecraft suffered no damage. "There isn't going to be an investigation. We know when it happened," he says. "There was a point when both of them were sitting on the same bench, and that has to have been it."

So, we send probes to an alien environment of which we know little and discover surprising differences which do not fit our preconceived model. This would seem to be an ideal opportunity for the application of the classical scientific method. After all, there are obvious signs of volcanic activity and suggestive evidence of the presence and action of water, both of which affect the concentrations of minerals in the one environment with which we are familiar, i.e., Earth. And the geological formations on Mars are widely varied: plains, mountains, rifts, canyons, craters, etc.

But the model requires that wind be the overwhelmingly predominate geological force on Mars. Ignore the implication that this postulated windswept homogeneous equilibrium would be reflected in a uniform dusty plain around the entire planet, at variance with the facts. Some clever scientist "discovers" that by simply switching the calibation curves of the instruments, the data better fits the model. Viola! Everyone knows NASA has made mistakes in the past. The technicians were very busy and stressed. The instruments were on the same bench at some time. That has to have been it -- we have proven it.

And so we have the modern "rational" scientific method: If the data does not fit the presumed model, simply invoke experimental incompetence.

Posted by J. D. Watson at March 6, 2005 10:34 PM


Posted by: Jed Roberts at March 6, 2005 10:47 PM

voila -- my french spelling was never very good.

Posted by: jd watson at March 7, 2005 12:01 AM

OJ, stop worrying about the state of science and start worrying about the Federal Budget. How about defund these clowns now.

Posted by: Robert Schwartz at March 7, 2005 12:16 AM

Actually, the scientific method would demand that we send two more probes to examine the first two to determine which one has which instrument. And that will happen, someday, mostly likely by some archeologist adding to some museum's collection or early Space Age artifacts. So these people who are depending on another NASA screwup for their theories to work had better hope that there really was a screwup. No wonder physisicts tend to vote for the Dems.

The real problem is that after forty odd years, these things are still being handmade one at a time by a modern equivalent to a medieval guild. If robots are so damn great, they should have been standardized by now such that any large university can buy one off the rack and send it themselves to confirm or deny the observations of earlier efforts, and the whole solar system should be crawling with them.

Posted by: Raoul Ortega at March 7, 2005 1:54 AM

Well, at least it seems the instruments are both using the same frames of reference (i.e., if one used metric units and one used English units, we would probably be hearing about giant molecules and strange densities right about now). Or that Mars is really two planets in one.

Posted by: jim hamlen at March 7, 2005 8:28 AM

J.D. - Experimental incompetence is actually the most common explanation for scientific results, although in the biomedical field fraud comes a close second. Prudent scientists know this, and re-check the experiment before going on to modeling and other costly theoretical activities.

This is why socialist science doesn't work well: there's no one to challenge results.

Posted by: pj at March 7, 2005 9:39 AM