Managing Geodata within the Site Selection Procedure
Bundesgesellschaft für Endlagerung mbH (BGE), Germany
This contribution outlines the components of the geodata management at the Site Selection Department of the BGE, responsible for the implementation of the German site selection procedure. Its first phase is based on existing data from federal and state authorities.
The BGE has implemented a comprehensive geodata management process. While incoming data and the related correspondence are stored in searchable archives, they are prepared in well-defined transfer workflows with implemented checks before usage. The data transfer from the geodata management department to the respective scientific departments includes checks for correct locations, coordinate systems, plausibility and completeness. Only standardized and officially recognized formats are used in the further analyses together with specialized software tools for GIS data, borehole data as well as geological models. The current data pool comprises 350 data deliveries from 41 institutions with approximately 1 Million data files.
A significant portion of the geological data are received in analog form from the federal geological and mining authorities. Since 2 years, the BGE puts significant efforts into digitizing these data in tight cooperation with qualified geo-consultant partners.
In accordance with the „Geologiedatengesetz“ (national law for usage of geological data), data relevant to the site selection procedure will be made publically available in order to fulfill obligations for public participation and the involvement of stakeholders and decision makers. The BGE strives to make data publically available as soon as the legal requirements are met and to continuously improve accessibility and user friendliness of its web-based data portal.
Artificial Intelligence in Geosciences: Time for a paradigm shift
ZALF and University of Potsdam, Germany
Geosciences face a dramatic increase of high quality data as well as of powerful artificial intelligence approaches. These new techniques, however, have mostly been limited to applications to pre-existing research questions and approaches, e.g. for parameterizing groundwater models. In hydrogeology, these paradigms are closely related to the previous approach of studying individual processes on small spatial and temporal scales and subsequent up-scaling, e.g., via conceptual or numerical models. However, that approach suffers from heterogeneities, interactions, and feedbacks between different processes which are inherent of natural systems, resulting in substantial uncertainties. Overcoming these scale issues is a major challenge both for science and for water resources management.
Modern artificial intelligence techniques, combined with dynamic system theory paradigms, pave the way to a different approach. They allow to extract meaningful information from extensive data sets directly at the scale of interest, e.g., for large regions. Thus constraints can be exploited that are not visible at small scales. An example will be presented, where the influence of heterogeneous land use on evapotranspiration, groundwater recharge and groundwater dynamics at the scale of 20,000 km2 was studied. It illustrates how science and water resources management can benefit a lot from exploring the range of now possible new scientific questions rather than from simple applications of artificial intelligence approaches in otherwise conventional studies.
An introduction to Landslide Susceptibility Assessment Tools - Project Manager Suite
Bundesanstalt für Geowissenschaften und Rohstoffe, Germany
Landslide Susceptibility Assessment Tools - Project Manager Suite (LSAT PMS), an open-source, user-friendly program written in Python developed and released at the Federal Institute for Geosciences and Natural Resources (BGR). Although initially developed to conduct landslide susceptibility analyses, LSAT PMS is applicable for all types of spatial analyses related to supervised binary classification. The first LSAT PMS release supports analysis workflows based on the weight of evidence, logistic regression, artificial neural network, and analytical hierarchy process. Solution tailored toolbox and the implemented data management environment allow efficient import, preprocessing, analysis and postprocessing of the data. The graphical user interface facilitates the intuitive exploratory work with the data and the models. Developing LSAT PMS, we focus on the practical assessment of uncertainties and model evaluation to better characterise the capabilities and limitations of implemented methods. Therefore, LSAT PMS offers different subsampling techniques and an evaluation tool to evaluate and compare models generated by different methods. Introducing LSAT PMS, we hope to provide easy access to state-of-the-art methods for the non-programming community supporting scientific principles of openness, knowledge integrity, and replicability. The standardised project framework of LSAT PMS allows an easy sharing of the data and model results among peers. With the utilisation of standard data formats, analysis results are transferable among all GIS for further processing and advanced visualisation. The software, corresponding comprehensive documentation, and a test dataset are ready for download on BGR’s home page and GitHub. LSAT PMS is subject to further development.
Understanding Natural Geomorphological Processes Through Artificial Intelligence and Crowdsourced Data
Academy for Mathematics, Science, and Engineering
As open source data becomes more ubiquitous, the involvement of citizen scientists has increased. The collection of large quantities of relevant data and respective labels through crowdsourcing on online platforms has yielded many exciting opportunities for machine learning applications. In geomorphology, multitemporal imagery, much of which is captured through crowdsourcing, is especially useful for training deep learning models for change detection in landscapes. This is relevant in terms of natural hazards that occur, including endogenous types like volcanoes and neotectonics, exogenous ones such as floods, karst collapses, sedimentation, erosion, tsunamis, and avalanches, as well as climate change or land use-induced hazards like permafrost and desertification. However, a challenge when harnessing crowdsourced imagery is the disorganized and “unclean” fashion in which it often presents itself. Cleaning data prior to training neural network-based computer vision models is key to success in any geomorphology change detection research. We discuss approaches such as manual techniques, image restoration and denoising, and image duplication reduction. The goal is to assimilate a diverse range of data collected from many sources to successfully train machine learning algorithms. In a broader sense, this research has the potential to save lives by detecting possibly destructive and dangerous geomorphological change, and to conserve environments that have been affected severely.
New phenomena in ESR spectra of iron ores from Kryvyi Rih deposit
Institute of Vocational Education, France
The electron spin resonance (ESR) spectra of iron ores from Kryvyi Rih deposit have been measured at two different temperatures: 295 K and 150 K. Two samples of ores were chosen for investigations: hematite ore of the Inguletsky combine of oxidized ores (sample 1) and hematite ore of the Novokryvorizky combine of oxidized ores (sample 2). The broad absorption lines with resonance field 1.546 kOe (sample1) and 1.453 kOe (sample 2) were observed at 295 K with values of the g-factor equal to 4.417 and 4.668 respectively. The amplitude of this line depends on the temperature. The ESR-signal amplitude increases with a reduction in temperature from 295 K to 150 K: for sample 1 by 21.6% and for sample 2 by 19.4%. The shift of the absorption line from 1.546 kOe to 1.456 kOe was observed with a reduction in temperature for sample 1 and from 1.453 kOe to 1.288 kOe was observed with a reduction in temperature for sample 2. The increase of the g-factor was observed with a reduction in temperature to 4.578 for sample 1 and to 5.390 for sample 2.
Peak Ring Magnetism: Rock- and mineral-magnetic properties of the Chicxulub impact crater
1Karlsruhe Institute of Technology, Germany; 2Utrecht University, Netherlands
Large impact structures on Earth like the Chicxulub in Mexico are characterized by magnetic highs but the magneto-mineralogical origin is still poorly constrained and impact-generated melt versus hydrothermal activity models are discussed. The IODP-ICDP expedition 364 drilled into the peak ring of the Chicxulub impact crater, which is characterized by a well-developed hydrothermal system. This system was active for up to 2 Ma, reaching temperatures of 350-450°C. The main goal of our study is the investigation and characterization of heat treatment on shocked magnetite, the most important magnetic mineral in the shocked granitoid basement, and impact lithologies from drill core M0077A.
In this study, we used a combination of microscopic, rock-magnetic, and paleomagnetic methods to investigate the potential post-shock temperature effects in magnetite. Our preliminary results suggest the presence of three types of magnetite. The first type found in the crystalline basement shows large fractured grains of pure magnetite, with scattered paleomagnetic directions. The second type consists of newly formed Al- and Mg- rich spinel, appearing in skeletal crystals at the uppermost impact melt layer, with stable 29r chron directions. A third type of magnetite is found throughout all lithologies in assemblage with sulphides, both interpreted of hydrothermal origin. We observe a general irreversibility in the temperature-dependent magnetic susceptibility (k-T curves) of the basement magnetite, and reversible k-T curves at close proximity with melt layers. We interpret this to indicate the hydrothermal system to not have reached annealing temperatures, in contrast with the slow-cooling, high-temperature deeper melt layers.
Temperature and frequency-dependent magnetic susceptibility parameters: improving the reliability of archaeointensity in burnt clay ceramics
1Posgrado en Ciencias de la Tierra, Instituto de Geofísica, Universidad Nacional Autónoma de México; 2Laboratorio de Paleomagnetismo, Instituto de Geofísica, Universidad Nacional Autónoma de México
Analysis of magnetic mineralogy alteration parameters has been used to understand mineralogy transformation at sample heating. While many studies used the reversibility of low-field magnetic susceptibility vs. temperature κ(T) curves in a qualitative sense to select material for paleomagnetic studies, few parameters have been developed to assess a quantitative description. This work aims to correlate the magnetic mineral properties deduced by susceptibility experiments of archaeological ceramics (burnt clay) during the heating steps of the Thellier-Thellier intensity method. Eight ceramics from a Mexican archaeological site were examined in cyclical experiments of κ(T) curves and susceptibility vs. frequency-dependence, respectively. We found that no degree in the reversibility of κ(T)-curves determine the successful samples for the archaeointensity estimation in advance, neither in single nor incremental temperature cycles. However, a complete analysis including more than seven cycles with an estimation of magnetic grain properties constrains the most useful samples for the archaeointensity experiment. We propose a new parameter (modIPT) to evaluate the apparent reversibility for cyclical κ(T)-curves. We found a significant correlation of this parameter with archaeointensity statistical values that infer remanence alteration or directional deviation produced by mineral transformations after heating steps. We realized that particular burnt clay material as archaeological ceramic samples are suitable to register a reliable geomagnetic intensity and consequently an accurate archaeological dating, even though similar selection filters used in preceding works could have underestimated these materials.
Utilising magnetic minerals to track and identify hydrocarbon migration pathways and source regions: a case study on the Beatrice Field, Inner Moray Firth, UK North Sea
Imperial College London, United Kingdom
Recent studies at Imperial College London have demonstrated that variations in magnetic mineralogy can be used to help track and quantify hydrocarbon migration. This work has built on past studies that identified strong magnetic anomalies associated with hydrocarbon accumulations, with several mechanisms suggested for their origin including: the influx of magnetic-mineral forming molecules creating new magnetic minerals; the formation of a reducing environment forcing chemical re-magnetisation of in situ minerals; and iron-forming bacteria biodegrading hydrocarbon organic matter. This study uses the connection between hydrocarbons and magnetic minerals to assist large-scale basin and petroleum systems modelling to answer a question that has existed since the 1970s – how did Beatrice Field in the Inner Moray Firth get charged?
Regional deformation imprints from anisotropy of magnetic susceptibility data – an example from the Raichur Schist Belt (Dharwar Craton, India)
1Department of Geology and Geophysics, Indian Institute of Technology (IIT), Kharagpur; 2Institute of Applied Geosciences, Karlsruhe Institute of Technology (KIT)
The Raichur Schist Belt (RSB) is a NW-SE trending late-Archaean greenstone belt that forms part of the supracrustal units lying over an older gneissic basement. Granites (ca 2.5 Ga) occur in the vicinity of the RSB. The metavolcanics and granites are both massive and lack a field foliation and/or lineation. To work out the time-relationship between emplacement, fabric development and regional deformation of the granite vis-à-vis metavolcanic rocks and regional deformation, we performed Anisotropy of Magnetic Susceptibility (AMS) studies. Mean magnetic susceptibility (Km) of the metavolcanics varies between 843 µSI and 57800 µSI units, while in the granite it is between 6.88 µSI and 45000 µSI units. Microstructural studies reveal that the rocks are deformed and AMS is mostly controlled by paramagnetic phases. Temperature-dependent magnetic susceptibility studies carried out so far establish that samples with Km>1260 µSI contain multidomain magnetite. In the metavolcanics, mean orientation of magnetic foliation is NNW-SSE; this is similar to D1/D2 regional fabric of Dharwar Craton. The magnetic lineation is doubly plunging (direction varying from NNW to SSE). This is a manifestation of D3 superposed on D1/D2 fabric in the metavolcanics of RSB inferred from magnetic fabric, the mesoscopic field evidence for which is lacking. Similar superposed deformation is also implied from the AMS data of granites. The region is replete with quartz veins and their orientation analysis with respect to the magnetic fabric is expected to provide further details about the kinematics of the rocks.
Exploring the preservation of greigite in hydrocarbon reservoirs using thermodynamic modelling
Imperial College London, United Kingdom
Previously, thermodynamic modelling has been used to predict the magnetic phases favoured under varying geochemical conditions at hydrocarbon seepage zones. Although greigite (Fe3S4) has been identified by magnetic experiments in the North Sea and Wytch Farm oilfields, it was not included in previous thermodynamic models. Multiple studies have outlined the conditions required for greigite preservation in nature: sulphur supply needs to be enough to form greigite but limited as to not proceed to form pyrite; total organic carbon content needs to be low as it can produce sulphides; there needs to be a high availability of reactive iron. This study uses thermodynamic modelling to help constrain the following: What is the optimum level of sulphur? How much available iron is required? At what temperatures is greigite stable? Answering these questions is the first step in determining how greigite can exist in hydrocarbon environments.