Application of Laser‐Induced Breakdown Spectroscopy and Hyperspectral Images for Direct Evaluation of Chemical Elemental Profiles of Coprolites

This study demonstrates the application of laser‐induced breakdown spectroscopy (LIBS) and hyperspectral imaging to the investigation of coprolite and fossil samples. Solid samples from the Permian (seven coprolites and one fossil), Cretaceous (one coprolite) and Oligo‐Miocene (two coprolites) periods were directly analysed, and emission spectra from 186 to 1042 nm were obtained in several areas covering coprolite/fossil and rock material. Initial exploratory analyses were performed using principal component analysis with the data set normalised by the norm (Euclidean norm = 1). After identification and selection of emission lines of eleven elements (Al, Ca, Cr, Fe, K, Mg, Mn, Na, Ni, P and Si), the signals were normalised again by the relative intensity of the selected element. Phosphorus was identified mainly in the coprolites, while K and Na were primarily found in the rock material. In several cases, there was a positive correlation between Ca and P. A sample from the Oligo‐Miocene series was also analysed using inductively coupled plasma‐optical emission spectrometry (ICP‐OES) (rock and coprolites were analysed separately). Based on the quantitative results from ICP‐OES, it was confirmed that the tendency was the same as that observed with the results obtained from LIBS directly in the solid sample.     

Improving the descendant relationship of cluster galaxies at z < 1

In this work, we study how to improve well-known techniques for detecting progenitors/descendants of galaxies, such as the NDpredict program, when applied to galaxies in clusters. The improvement of this particular method is based on the use of the red sequence of galaxies in those environments. Objects close to the red sequence in the color and magnitude diagram are more likely to belong to the cluster. This defines a probability scale which is then combined with the one generated by NDpredict. This procedure is optimized for the study of galaxies in clusters over different epochs. Our main result is that, for a sample composed of $120$ clusters, with masses greater than ${10}^{13.25}{M}_{\odot }$, selected from the IllustrisTNG simulations (namely, the TNG100 runs). In $99\%$ of the cases (i.e., $119$ systems), we obtain better performance with the red sequence method in comparison to the original NDpredict, and the average gain obtained is $28\%$ in the identification of descendants for this sample of cluster galaxies.