Logo ENSG réduit

Digital in the heart
of the territories

Logo facebookLogo twitterLogo linkedInFR


An (almost) automated process to track the Martians dunes.

Arthur, Engineer 2nd year, 2017-2018

The California Institute of Technology (Caltech).


Currently, measuring the motion of dunes on Mars is done manually. It is a painstaking procedure that requires a lot of time and precision, so it is impossible to generalize this to the whole planet. Yet, the phenomena behind these displacements (surface winds, saltation...) are the main motors of the evolution of the Martian reliefs. If we want to understand how the Red Planet is shaped by these atmospheric phenomena without having a plethora of meteorological stations scattered on its surface, it is necessary for us to develop a technique to be able to track the dunes and characterize their displacement over several years and on the scale of the planet.
During my internship, I had to find a way to measure – almost automatically (i.e. without the operator repeating over and over the same long and complicated protocol for each of the Martians dunes and for each new acquisition) – the movement of the dunes on Mars, using remote sensing and image processing skills and techniques. Once our protocol was found, it still had to be implemented, while keeping in mind that it had to remain accessible (i.e. usable by a geologist without particular computing skills for instance). it is for this purpose that I created a Python package, named ac.GPS (ArthurCoqué.GetPreciseShifts) allowing to easily perform this task (as well as other related functions).

Planetology, Remote sensing, Image processing, Python.

Updated the 14 December 2018