Testing the X-ray computed microtomography on microfossil identification: An example from Sergipe-Alagoas Basin, Brazil

Abstract

Micropaleontology is crucial branch of Earth Sciences, with a pivotal role in the success of the oil and gas industry over the decades. This sector of paleontology is based on the taxonomical description of microfossils, which encompass fossils with size variation from 0.001 mm to 1 m. Normally these microorganisms have a high rate on preservation, thus they are widely accepted as reliable evidence to infer about paleodepositional settings and paleoclimate change. Furthermore, microfossils are critical to understanding the evolution through time, being many of them recognized as index fossil and providing useful biozones to correlation. Since 1950, microfossils have been taxonomically described mainly by stereomicroscopes. However, as the technology progress, traditional ways to study these organisms are challenged and improved by nondestructive three-dimensional imaging techniques, as X-ray tomographic microscopy (SRXTM), three-dimensional X-ray microscopy (3D-XRM), the X-ray computed microtomography (microCT) and X-ray computed nanotomography (nanoCT). Recently, one of the most compelling research areas in micropaleontology is the desire to automatize and enhance the details of systematic classification. Thereby, an increasing number of researches have applied the high-resolution X-ray analysis aiming to improve the morphological, taxonomic and taphonomic examination. Most of them have detailed the enhancement provided by the technique when compared with the standard microscopes, and raised questions about the traditional characters used on the microfossil systematic taxonomy. Nevertheless, even with the surpassing detail on microfossil characterization, the application of microCT has been hampered by the costs and sometimes by the needs of specific computer skill. Thus, this research has evaluated the use of microCT as the technique to classify a random bulk of microfossil (comprising foraminifers, ostracods, radiolarians, gastropods and echinoderms) with no further software treatment. No standard microscope analysis was performed. Despite chemical composition of microfossils, most of the specimens morphology, especially internal structures, have been easily acquired and analyzed. 96% of the microfossils of the dataset were identified at least on genus level. Irrespective of the deepness of detail, when considering the overall taxonomic identification, the microCT seems to be effective as the standard microscope. Nonetheless, when problematic specimens are evaluated, the microCT seems to be a more reliable and practical tool than other methods as Scanning Electron Microscopy. Thus, the technique can be used solo or as a complementary method to the stereomicroscope. Additionally, the high-resolution has the potential to lead to the expected automatized micropaleontology, since they can provide numerous images in several planes. This may create a strong database necessary to machine learning and computer identification.