Automations in Chemostratigraphy: Toward Robust Chemical Data Analysis and Interpretation

Elemental chemostratigraphy has become an established stratigraphic correlation technique over the last 15 years. Geochemical data are generated from rock samples (e.g., ditch cuttings, cores or hand specimens) for up to c. 50 elements in the range Na-U in the periodic table using various analytical techniques. The data are commonly displayed and interpreted as ratios, indices and proxy values in profile form against depth. The large number of possible combinations between the determined elements (more than a thousand combinations), makes it a time-consuming effort to identify meaningful variations that resulted in correlative chemostratigraphic boundaries and zones between wells. The large number of combination means that 30-40% of the information is not used for the correlations that maybe crucial to understand the geological processes. Automation and artificial intelligence (AI) are envisaged as likely solutions to this challenge. Statistical and machine learning techniques are tested as a first step to automate and establish a workflow to define (chemo-) stratigraphic boundaries, and to identify geological formations. The workflow commences with a quality check of the input data and then with principle component analysis (PCA) as a multivariate statistical method. PCA is used to minimize the number of elements/ratios plotted in profile form, whilst simultaneously identifying multidimensional relationships between them. A statistical boundary picking method is then applied define chemostratigraphic zones, for which reliability is determined utilizing quartile analysis, which tests the overlap of chemical signals across these statistical boundaries. Machine learning via discriminant function analysis (DFA) has been developed to predict the placement of correlative boundaries between adjacent sections/wells. The proposed workflow has been tested on various geological formations and areas in Saudi Arabia. The chemostratigraphic correlations proposed using this workflow broadly correspond to those defined in the standard workflow by experienced chemostratigraphers, while interpretation times and subjectivity are reduced. While machine learning via DFA is currently further researched, early results of the workflow are very encouraging. A user-friendly software application with workflows and algorithms ultimately leading to automation of the processes is under development.

Enhancing Stratigraphic Framework Consistency Using Spectral Gamma-Ray Data

Stratigraphic correlation is crucial for reservoir characterization; therefore, it requires more advanced methods and techniques to reduce the stratigraphic correlation uncertainty, especially when variation in lateral facies is high. The studied formations from bottom to top consist of fluvial to marginal marine X Formation, shallow marine Y Formation, and fluvial distributary channels to estuarine Z Formation. Spectral gamma-ray logs give additional consistent information on lithological composition that can support identification of boundary between formations within the stratigraphic framework. Wells with a full section of Y Formation, core, palynology, and spectral gamma-ray were selected as key wells. The top and base of the Y Formation were picked using conventional logs refined by a thorium/potassium (Th/K) ratio log and cross plot with core and palynology data as validations. The internal Y Formation markers were also picked with the aid of the Th/K cross plots. The top picking criteria from the key wells was implemented to the rest of the wells across the field with consistency. The uniform low Th/K ratio log (<3.5) across the Y Formation indicates illite as the dominant clay type, confirmed by X-ray diffraction (XRD) data with an average of more than 80%. The character is consistent with the interpreted depositional environment. This character makes the Y Formation stand out from the overlying Z and the underlying X formations. The change from X to Y Formation is defined by the decrease of the Th/K ratio log, from high kaolinite content to illite dominated environment. Inversely, the top of the Y Formation (base of Z) is indicated by the increase of the Th/K ratio log moving from shallow marine Y Formation to the fluvial-influenced Z Formation. The Th/K cross plot indicates different clusters amongst the studied formations and the internal Y zonation. The X Formation is located in the high Th and low K area where kaolinite is predominant, related to fluvial environment. The case is similar for the Z Formation but with more influence of mixed-clay type. The Y Formation shows clear clustering along the mixed-clay and illite window. Internal Y zonation displays, from bottom to top, an increasing K value within the clusters. This method provides a semi-quantitative interpretation to define the studied formations boundaries and the Y Formation internal zonation. This study has increased the consistency of the studied formations’ stratigraphic and structural framework. This consistency has, in turn, fine-tuned the structural framework and aided field development through better geosteering and lateral well placements. These results are a valuable starting point to refine and extend the work to other areas.

The Invention and Application of Typical Points-Based Well Log Correlation and Picking Up Technology

This paper proposes a new method called typical points-based well log correlation and picking up technology and provides several related application examples based on this method. The new method firstly determines representative extreme points, typical or characteristic points by analyzing the characteristics of logging curves and lithology of different wells, which are generally representative points with special geological significance, including the points with the best physical properties or tight points. (For example, the maximum flooding surface or exposed surface in a sedimentary cycle, etc.). On the basis of these characteristic points, we carry out stratigraphic correlation and tracking between wells to obtain a data set of a series of characteristic points. From the same characteristic point, all points have the same or similar petrophysical properties, and the logging curve values of these characteristic points are extracted. And then analyze the change trend, distribution characteristics and the internal relationship of the parameters of the data set of each feature point. Based on the data set obtained from the method above, we extended it to the following application areas: 1) Through mathematical theoretical models, two free water level distribution modes and their determination workflows were established, including horizontal and tilted free water levels. 2) Perform data quality analysis and control, especially logging data analysis. 3) Exploratory application in the standardization of logging curves.4) Application in dynamic performance analysis The new method is developed on the traditional stratigraphic correlation method and stratal slicing method (Zeng Hongliu, 1998) and then used for well log data extraction and analysis. It is a practical means and technique for geological analysis. The application effect shows that the it is reliable, convenient and practical.

Early–Middle Pleistocene Magnetostratigraphic and Rock Magnetic Records of the Dolynske Section (Lower Danube, Ukraine) and Their Application to the Correlation of Loess–Palaeosol Sequences in Eastern and South-Eastern Europe

We present new palaeomagnetic and rock magnetic results with a stratigraphic interpretation of the late Early–Middle Pleistocene deposits exposed on the left bank of the River Danube at Dolynske, southern Ukraine. A thick succession of water-lain facies is succeeded by reddish-brown clayey soils, topped by a high-resolution loess–palaeosol sequence. These constitute one of the most complete recently discovered palaeoclimate archives in the Lower Danube Basin. The suggested stratigraphy is based on the position of the Matuyama–Brunhes boundary, rock magnetic, palaeopedological and sedimentological proxies, and it is confidently correlated with other loess records in the region (Roksolany and Kurortne), as well as with the marine isotope stratigraphy. The magnetic susceptibility records and palaeosol characteristics at Dolynske show an outstanding pattern that is transitional between eastern and south-eastern European loess records. Our data confirm that the well-developed S4 soil unit in Ukraine, and S5 units in Romania, Bulgaria and Serbia, correlate with the warm MIS 11. Furthermore, we suggest the correlation of rubified S6 palaeosols in Romania and Bulgaria and the V-S7–V-S8 double palaeosol in Serbia with S6 in Ukraine, a strong Mediterranean-type palaeosol which corresponds to MIS 15. Our new results do not support the hypothesis of a large magnetic lock-in depth like that previously interpreted for the Danube loess, and they prove that the Matuyama–Brunhes boundary is located within the palaeosol unit corresponding to MIS 19. The proposed stratigraphic correlation scheme may serve as a potential basis for further regional and global Pleistocene climatic reconstructions.

Semantic segmentation of vertebrate microfossils from computed tomography data using a deep learning approach

Vertebrate microfossils have broad applications in evolutionary biology and stratigraphy research areas such as the evolution of hard tissues and stratigraphic correlation. Classification is one of the basic tasks of vertebrate microfossil studies. With the development of techniques for virtual paleontology, vertebrate microfossils can be classified efficiently based on 3D volumes. The semantic segmentation of different fossils and their classes from CT data is a crucial step in the reconstruction of their 3D volumes. Traditional segmentation methods adopt thresholding combined with manual labeling, which is a time-consuming process. Our study proposes a deep-learning-based (DL-based) semantic segmentation method for vertebrate microfossils from CT data. To assess the performance of the method, we conducted extensive experiments on nearly 500 fish microfossils. The results show that the intersection over union (IoU) performance metric arrived at least 94.39 %, meeting the semantic segmentation requirements of paleontologists. We expect that the DL-based method could also be applied to other fossils from CT data with good performance.

Multi-scale virtual field experience: sedimentology and stratigraphy of Grand Ledge, Michigan, USA

Field experiences are a critical component of undergraduate geoscience education; however, traditional onsite field experiences are not always practical due to accessibility, and the popularity of alternative modes of learning in higher education is increasing. One way to support student access to field experiences is through virtual field trips, implemented either independently or in conjunction with in-person field trips. We created a virtual field trip (VFT) to Grand Ledge, a regionally important suite of sedimentary outcrops in central lower Michigan, USA. This VFT undertakes all stages of a field project, from question development and detailed observation through data collection to interpretation. The VFT was implemented in undergraduate sedimentation and stratigraphy courses at two different liberal arts institutions, with one version of the VFT conducted in-person and the other online. The VFT was presented from a locally hosted website and distributed through an online learning platform. Students completed a series of activities using field data in the form of outcrop photos, virtual 3D models of outcrops and hand samples, and photos of thin sections. Student products included annotated field notes, a stratigraphic column, a collaborative stratigraphic correlation, and a final written reflection. VFT assessment demonstrated that students successfully achieved the inquiry-oriented student learning outcomes, and student reflection responses provide anecdotal evidence that the field experience was comparable to field geology onsite. This VFT is an example of successful student learning in an upper-level sedimentation and stratigraphy course via virtual field experience with an emphasis on local geology.

OSL Chronology of the Siling Co Paleolithic Site in Central Tibetan Plateau

When and how was the Tibetan Plateau (TP), one of the least habitable regions on Earth, occupied by humans are important questions in the research of human evolution. Among tens of Paleolithic archaeological sites discovered over the past decades, only five are considered coeval with or older than the Last Glacial Maximum (LGM, ∼27–19 ka). As one of them, the Siling Co site in the central TP was previously announced to be ∼40–30 ka based on radiocarbon dating and stratigraphic correlation. Given the loose chronological constraint in previous studies, we here re-examined the chronology of the Siling Co site with the optically stimulated luminescence (OSL) dating technique. Four sections from the paleo-shoreline at an elevation of ∼4,600 m in southeastern Siling Co were investigated, with stone artifacts found from the ground surface. Dating results of nine samples delineated the age of ∼4,600 m paleo-shoreline to be ∼10–7 ka (∼8.54 ± 0.21 ka in average). This age indicates that the Siling Co site is not earlier than the early Holocene, much younger than the former age. The revised age of the Siling Co site is consistent with the wet and humid climate conditions on the TP during the early Holocene.

Aplicação da perfilagem geofísica na correlação estratigráfica de poços / Application of geophysical logging in the stratigraphic correlation of wells

A correlação estratigráfica de poços é de suma importância na indústria do petróleo. Ela tem como objetivo a determinação da continuidade lateral das rochas. A partir dela é possível definir estratégias de explotação de um campo petrolífero, bem como a continuidade do reservatório. Porém, a técnica da correlação estratigráfica não é uma tarefa fácil, e está sujeita a interpretações equivocadas, devido a variabilidade geológica e à ambiguidade das respostas das ferramentas de perfilagem. Com objetivo de melhor compreender e definir a continuidade lateral das camadas e dos reservatórios do Campo de Namorado, Bacia de Campos – RJ. Para isso foram selecionados três poços (NA45D, NA52D e NA49D) com dados de perfis de Raios Gama, Resistividade, Porosidade Neutrão e Densidade. Os perfis foram gerados através do software LogPlot e posteriormente interpretados. O intervalo do reservatório foi identificado em cada poço e correlacionado, o que possibilitou observar a continuidade do reservatório do Campo de Namorado e sua heterogeneidade.

Enhancement of Indistinct Sequence Stratigraphic Correlations Using Geochemical Signatures: An Example from the Paleozoic Successions Saudi Arabia

The Paleozoic sequence in Kingdom of Saudi Arabia (KSA) represents a thick succession of carbonate, shale, and siltstone/sandstone lithologies that were deposited in a range of environments. This succession hosts abundant unconventional reservoirs with huge amount of natural gas reserves. However, similarity in lithologies and poor biostratigraphic constrains are making facies analysis, stratigraphic correlation, and geosteering applications problematic. In this study, we documented the use of an automated statistical protocol for the identification of chemofacies and correlatable chemozones within highly homogenous formations.