Abstract Barchan dunes usually appear under the action of one‐directional fluid flows, being found on Earth, Mars, and other celestial bodies; however, although ubiquitous in nature, understanding dune dynamics at the grain scale is challenging due to the vast number of grains involved. Here, based on subaqueous experiments using a high‐speed camera, discrete numerical computations solving the motion at the grain scale, and a special training of a convolutional neural network, we show that it is, in fact, possible to estimate the resultant force acting on the grains of a barchan dune by using images. This procedure opens new possibilities for measuring the resultant force not only on the grains of a dune, but also on relatively small elements that are imaged over time, such as rocks, boulders, rovers, and human‐built constructions photographed by satellites on terrestrial and Martian landscapes.

Read original article

Read original article