We present a pipeline for geomorphological analysis that uses structure from motion (SfM) and deep learning on close-range aerial imagery to estimate spatial distributions of rock traits (diameter, size, and orientation) along a tectonic fault scarp. Unpiloted aircraft systems (UAS) have enabled acquisition of high-resolution imagery at close range, revolutionizing domains such as infrastructure inspection, precision agriculture, and disaster response. Our pipeline leverages UAS-based imagery to help scientists gain a better understanding of tectonic surface processes. We start by using SfM on aerial imagery to produce georeferenced orthomosaics and digital elevation models (DEM), then a human expert annotates rocks on a set of image tiles sampled from the orthomosaics. These annotations are used to train a deep neural network to detect and segment individual rocks in the whole site. This pipeline automatically extracts semantic information (rock boundaries) on large volumes of unlabeled, high-resolution aerial imagery, which allows subsequent structural analysis and shape descriptors to result in estimates of rock diameter, size and orientation. We present results of two experiments conducted along a fault scarp in the Volcanic Tablelands near Bishop, California. We conducted the first experiment with a hexrotor and a multispectral camera to produce a DEM and five spectral orthomosaics in red, green, blue, red edge (RE), and near infrared (NIR). We then trained deep neural networks with different input channel combinations to study the most effective learning method for inference. In the second experiment, we deployed a DJI Phantom 4 Pro equipped with an RGB camera, and focused on the spatial difference of rock-trait histograms in a larger area. Although presented in the context of geology, our pipeline can be extended to a variety of geomorphological analysis tasks in other domains.

Angular distribution measurements of $^2$H($^7$Be,$^7$Be)$^2$H and $^2$H($^7$Be,$^8$B)$n$ reactions at $E_{c.m.}\sim$~4.5 MeV were performed to extract the astrophysical $S_{17}(0)$ factor using the asymptotic normalization coefficient (ANC) method. For this purpose a pure, low emittance $^7$Be beam was separated from the primary $^7$Li beam by a recoil mass spectrometer operated in a novel mode. A beam stopper at 0$^{\circ}$ allowed the use of a higher $^7$Be beam intensity. Measurement of the elastic scattering in the entrance channel using kinematic coincidence, facilitated the determination of the optical model parameters needed for the analysis of the transfer data. The present measurement significantly reduces errors in the extracted $^7$Be(p,$\gamma$) cross section using the ANC method. We get $S_{17}$~(0)~=~20.7~$\pm$~2.4 eV~b.