Mars mix up and data quality

Do you trust the temperature reading on your thermometer? How important is it to us that equipment records the ‘correct data’?

You may remember a few years ago (six and a half) that NASA landed a couple of robots on Mars – Opportunity and Spirit.

Image kindly reproduced according to the licence at: http://en.wikipedia.org/wiki/File:NASA_Mars_Rover.jpg

The rovers were scheduled to last for a 90 day misson – well, amazingly, they are still working – although Spirit is stuck in sand and hopes are fading for getting it out and communicating with NASA. Opportunity, however, is still trundling away.

Both rovers are equipped with an X-ray spectrometer, to analyse the chemical composition of rocks on Mars. However, for the first few months of their missions, both rovers were reporting slightly different concentrations of elements in the soil.

The results should have been the same as on a windswept planet such as Mars, the concentrations of the elements should have evened out and been the same.

So what was happening? It took a sharp eyed scientist to realise that the rovers have been equipped with the others spectrometer. In other words, Spirit has Opportunity’s and Opportunity had Spirit’s spectrometer.

Why did this make a difference?

The spectrometers were naturally imperfect and before use needed to be calibrated. Each spectrometer needed a slightly different calibration.

NASA engineers accidentally place Spirit’s correctly calibrated spectrometer into Opportunity and vice versa. The robots and NASA scientists assumed that both rovers had the correct instrument – and so when they were used they fed back incorrect results.

Once the mistake was spotted a quick software update fixed the problem.

So next time you use a piece of equipment to make a measurement, ask yourself the question, how do I know that this instrument is giving me the correct data? :mrgreen:

Article adapted from an article in New Scientist magazine, page 18, 5th March 2005 titled ‘Twin Rovers in instrument mix-up’.

No Comments Yet.

Leave a Reply

Your email address will not be published. Required fields are marked *


*