Despite the closeness between the Moon and the Earth, no global lunar map had been made until just two days ago.
On the 16th of November the dataset we had been waiting for since the Apollo era, according to Mark Robinson, Principal Investigator of the LROC, was revealed. The LROC (Lunar Recconnaissance Orbiter Camera) is the responsible for taking the pictures. This instrument belongs to a robotic spacecraft orbiting the Moon since 2009 with the aim of identifying safe landing sites, locating potential resources on the Moon, characterizing the radiation environment, and demonstrating new technology.
The LROC is made up of three cameras: two narrow and one wide angular one. A very similar camera to this last (WAC) is being used in another parallel programm around Mars.
The camera orbits at an average altitude of 50km and has a pixel scale of about 75 meters, so a WAC image swath is 70km wide around the ground-track, so it nearly covers the entire lunar surface in around one month. However we don't obtain the same images every month, but with tocks reflecting light under different conditions. This collection of stereo images are the ones that -after being treated- lead as to the final model. 69000 stereo images are need to get it. In spite of this huge amount of information, there are presistent shadows near the poles, but the spacecraft includes a laser altimeter (LOLA) that provides a precise topographic reconstruction since the spacecraft orbits converge at the poles therefore the "pole holes" can be filled.
The model is called GLD100 and covers 98,2% of the lunar surface and it was obtained this way:
The WAC stereo images arecompared one against another by pattern-matching a moving box of pixels until the best fit was found between two images with different viewing angles. Best fit pixel positions are combined with the LRO orbit position and the WAC viewing angles to define two 3D rays (lines of sight). The intersection point of these rays defines the location and the elevation of the point on the surface. Since the correlation box is bigger than 100 meters, surface details at the 100-meter scale are not fully resolved in a single stereo pair. However, each 100 meter square has an average of 26 stereo points within it , which helps to sharpen the elevation estimate. The accuracy of the elevations is estimated to be about 10 to 20 meters. Anyway, this map was built from the first year of stereo imaging, but there is already data corresponding to another year, what will make possible a more accurate model.
This project is related to our CanSat secondary mission, also consisting on creating a 3D map from previously treated images.
Via: NASA
No comments:
Post a Comment