PAGE 12
The Thermal Emission Imaging System (THEMIS) is an instrument aboard the Mars Odyssey spacecraft, that contains an infrared imager, launched from Kennedy Space Center on April 7th 2001.
THEMIS is capable of seeing through small amounts of dust. Longwave infrared imagers have a limited ability to penetrate underground, by detecting surface "hot spots" conducting heat upward from below. But they can only sense a limited depth, unless the material it is seeing through is a layer of fluffy micron sized dust. This type of dust would essentially be transparent to the THEMIS wavebands. Ice is also transparent to the THEMIS imager.
The Mars Orbiter Laser Altimeter, (MOLA), an instrument aboard the Mars Global Surveyor, was able to determine that in fact the surface of Cydonia does in fact consist of dust the size of only 1.06 microns.
When Enterprise Mission researcher Keith Laney first took a look at the new THEMIS data on July 25th, 2002, he was largely unimpressed. Saying that the image "sucked." It was not the high resolution false color images that he had been expecting from THEMIS.
Keith Laney is a digital imaging and and software specialist, and MOC image processor for the NASA-Ames' MOC MER2003 Landing Sites Project.
After some prodding, Keith eventually returned to the images he had downloaded form the THEMIS web site. After some rudimentary enhancements, he noticed some "blocky noise" in the images, and decided to try to to remove that noise. Nothing he tried could seem to remove the noise, nor could anyone find the same kind of blocky noise in other published THEMIS data of other areas on Mars. In fact this noise seemed to be regularly unique to the Cydonia THEMIS Infrared Images.
Keith then acquired the ENVI 3.5 and IDL software. This package is touted as the commercial multi-spectral imaging tool. By using this software, he again tried to remove the noise, and was unsuccessful. It was at this point that principal investigator Richard C. Hoagland decided to suggest to Keith that he tried a well known technique called "noise-averaging" which has been used for decades in astronomical photography known as luminance layering.
Things immediately got very interesting. What they found, was that instead of the noise going away, it actually became stronger. They realized, that what they thought of as the noise, may actually be the signal. Keith then began to amplify the noise pattern in his ENVI software processing. What they found, is pictured in the images below, mile upon mile of a lost cityscape, buried under ice and dust:
Now, some will say that the researchers are merely enhancing the noise. But what's interesting about this noise, is that the pattern does not follow the scan lines of the imager. Instead, the pattern follows the true North and South of Cydonia, exactly as terrestrial cities often do. Also, there are a wide variety of individual "structures" which do not conform to this noise pattern at all.
