Geophysical Imaging and Monitoring
Completed thesis projects
-
Full Waveform Inversion Using a Gaussian Process Emulator
2024 |
M.Sc. Applied Geosciences
Abstract
Full waveform inversion (FWI) is a currently promising method for imaging rock physical parameters from seismic wave data. The general workflow is to simulate the wave propagation iteratively, until the model parameters that produce the most similar simulated data to the original one are found. However, simulation requires expensive computing resources and is a lengthy process. Various research efforts have been attempted to speed up and optimize the FWI workflow. One way is to use a surrogate model or emulator, which is done in this work. Gaussian Process (GP) is a regression method based on the combined distribution of random variables from certain kernel functions. The expected output is the mean distribution of the posterior random function and the confidence interval. The regression model from GP can be a substitute model to reduce the number of iterations and predicting the parameter selection on the simulation results. In this thesis, the parameters chosen are P- and S-wave velocity, rock density, and quality factor. The expected output is the root mean square error. The seismic wave propagation simulator was NEXD for 1-D and 2D, while for 3-D it used SPECFEM-3D Cartesian. The GP emulator used the Psimpy Python library package. The methodology and algorithms were tested using synthetic models in 1-D and 2-D to ease the trial-and-error process. Workflows that are assured to be appropriate then applied to the ultrasound lab measurement data on four different homogeneous core samples (aluminium, marble, sandstone, and polyoxymethylene) and simulated in 3-D. The results of synthetic data trials were very accurate in determining model parameters that explain the data. In the core data, the less attenuated medium shown satisfactory results, but the ones with a high attenuation experienced challenges and required more iterations.
-
Optimale Versuchsplanung für die langfristige refraktionsseismische Überwachung der aktiven Auftauschicht am Schilthorn, Schweizer Alpen
2023 |
B.Sc. Applied Geosciences
Note: Janna Blaume continued on a geophysical path within the IDEA League Joint Master program.Abstract
In Rahmen dieser Arbeit wird die optimale Versuchsplanung für die langfristige refraktionsseismische Überwachung der aktiven Auftauschicht am Schilthorn in den Schweizer Alpen untersucht. Angewendet werden folgende Methoden: Ein analytisches Zweischichtmodell zur Bestimmung der optimalen Geophonabstände in Abhängigkeit der Zunahme der aktiven Auftauschicht sowie die Validierung der Ergebnisse durch Modellierung von synthetischen sowie Felddaten. Die Felddaten wurden von PERMOS aus den Jahren 2008- 2022 erhoben. Folgende Ergebnisse wurden generiert: Je mächtiger die ALT wird, desto niedriger ist der rrms und desto größer sind die optimalen Geophonabstände. Wenn das Schilthorn auch in Zukunft eine gute Auflösung des Permafrostes bieten soll, dann muss die Messgeometrie angepasst werden.
-
Minimum entropy constraints for 3D structurally-coupled joint inversion of near-surface geophysical data acquired at the Rockeskyller Kopf, Germany
2023 |
M.Sc. Applied Geophysics
Note: Anton Ziegon was the first official GIM graduate.Abstract
Geophysical methods are widely used to gather information about the subsurface as they are nonintrusive and comparably cheap during acquisition, however, the solution to the geophysical inverse problem is inherently non-unique which introduces considerable uncertainties. Therefore, independently acquired geophysical data sets can be jointly inverted to reduce ambiguities in the resulting multi-physical subsurface images. Zhdanov et al. (2022) introduce a novel cooperative inversion approach using joint minimum entropy constraints in the regularization term of the objective functionals to create more consistent multi-physical images with sharper boundaries. Here, this approach is implemented in an open-source software and its applicability on electrical resistivity tomography (ERT), seismic refraction tomography (SRT) and magnetic data is investigated. A synthetic 2D ERT and SRT data study is used to demonstrate the approach and to investigate the influence of the governing parameters. The findings showcase the advantage of the joint minimum entropy (JME) stabilizer over separate, conventional smoothness-constrained inversions. The method is then used to analyze field data from Rockeskyller Kopf, Germany. 3D ERT and magnetic data is combined and results confirm the expected volcanic diatreme structure with improved details. The multi-physical images of both methods are consistent in some regions as similar boundaries are produced in the resulting models, which have been lacking in previous studies. Because of its sensitivity to hydrologic conditions in the subsurface, observations suggest that the ERT method senses different structures than the magnetic method. However, these structures in the ERT result do not seem to be enforced on the magnetic susceptibility distribution, showcasing the flexibility of the approach. Both investigations outline the importance of a suitable parameter and reference model selection for the performance of the approach and suggest careful parameter tests prior to the joint inversion. With proper settings, the JME inversion is a promising tool for geophysical imaging, however, this thesis also lists some objectives for future studies and additional research to explore and optimize the method.
-
Imaging of active layer characteristics through quasi-3D inversion of frequency-domain electromagnetic soundings
2022 |
M.Sc. Applied Geophysics
Note: This thesis was supervised together with Dr. Sebastian Uhlemann (Lawrence Berkeley National Lab) and conducted during Florian's time as substitute professor of the Institute for Applied Geophysics and Geothermal Energy (now Computational Geoscience, Geothermics and Reservoir Geophysics).Abstract
The active layer thickness has become an important indicator in climate change research as permafrost degradation has long been documented. The thawing of permafrost causes the release of greenhouse gases accelerating Arctic warming. Monitoring and quantifying spatial and temporal changes of the active layer are challenging but crucial for reliable climate projections. Geophysical methods offer a non-invasive investigation of electrical properties and their distribution in permafrost areas, revealing phase transitions from water to ice. Subsurface electrical resistivity images can be obtained through inversion of electromagnetic data, yet are inherently ambiguous because of the ill-posed nature of the inverse problem. Since regularization methods offer the possibility to stabilize the inversion, lateral and spatial constraints are incorporated in the inversion algorithm to produce quasi-2D and quasi-3D subsurface models. The developed methodology is evaluated based on synthetic data sets to determine suitable inversion parameters, which are subsequently applied to a field example from the Seward Peninsula, Alaska. Laterally constrained inversion methods based on a fewlayer starting model succeed in resolving sharp interfaces in quasi-layered environments. In more complex settings minimum-structure models can retrieve accurate subsurface representations leveraging on vertical and horizontal smoothness constraints. Enforcing lateral and spatial consistency between neighboring soundings thereby yields a similar degree of model smoothness. The inverted field data confirms the conclusions drawn from the synthetic study, as meaningful three-layered models with regard to electrical resistivities are recovered, indicating resistive snow overlying the conductive active layer and highly resistive permafrost. However, the inversion results imply that the snow layer has a significant effect on the predicted model. The implemented constraints help in reducing the ambiguity of the models, but uncertainties introduced by limited data availability cannot be overcome. The potential of adopting spatial and lateral constraints to the inversion is shown, although it becomes evident that additional a priori information needs to be integrated in the objective function in order to comprehensively image the active layer.
-
Automated assessment of slope stability - application of machine learning to ERT monitoring data and integration with geomechanical models
2021 |
M.Sc. Applied Geophysics
Note: This thesis was supervised together with Dr. Sebastian Uhlemann (Lawrence Berkeley National Lab) and conducted during Florian's time as substitute professor of the Institute for Applied Geophysics and Geothermal Energy (now Computational Geoscience, Geothermics and Reservoir Geophysics). -
Process-based geophysical imaging of permafrost degradation: Is it getting wetter or drier?
2021 |
M.Sc. Applied Geophysics
Note: This thesis was supervised together with Prof. Christian Hauck (University of Fribourg) and conducted during Florian's time as substitute professor of the Institute for Applied Geophysics and Geothermal Energy (now Computational Geoscience, Geothermics and Reservoir Geophysics). -
Geoelectrical imaging of subsurface discontinuities and heterogeneities using low-dimensional parameterizations
2021 |
M.Sc. Applied Geosciences
Note: This thesis was supervised together with Prof. Florian Wellmann and conducted during Florian Wagner's time as substitute professor of the Institute for Applied Geophysics and Geothermal Energy (now Computational Geoscience, Geothermics and Reservoir Geophysics). -
Improved quantitative imaging of alpine permafrost evolution through time-lapse joint inversion of seismic refraction and electrical resistivity data
2020 |
M.Sc. Applied Geophysics
Note: Johanna Klahold received the Heitfeld award for her work. This thesis was supervised together with Prof. Christian Hauck (University of Fribourg) and conducted during Florian's time as substitute professor of the Institute for Applied Geophysics and Geothermal Energy (now Computational Geoscience, Geothermics and Reservoir Geophysics).Abstract
Quantitative estimation of pore fractions filled with liquid water, ice and air is one of the prerequisites in many permafrost studies and forms the basis for a process-based understanding of permafrost and the hazard potential of its degradation in the context of global warming. The volumetric ice content is however difficult to retrieve, since standard borehole temperature monitoring is unable to provide any ice content estimation. Geophysical methods offer opportunities to image distributions of permafrost constituents in a non-invasive manner. A petrophysical joint inversion was recently developed to determine volumetric water, ice, air and rock contents from seismic refraction and electrical resistivity data. This approach benefits from the complementary sensitivities of seismic and electrical data to the phase change between ice and liquid water. A remaining weak point was the unresolved petrophysical ambiguity between ice and rock matrix. Within this study, the petrophysical joint inversion approach is extended along the time axis and respective temporal constraints are introduced. If the porosity (and other time-invariant properties like pore water resistivity or Archie exponents) can be assumed invariant over the considered time period, water, ice and air contents can be estimated together with a temporally constant (but spatially variable) porosity distribution. It is hypothesized that including multiple time steps in the inverse problem increases the ratio of data and parameters and leads to a more accurate distinction between ice and rock content. Based on a synthetic example and a field data set from an Alpine permafrost site (Schilthorn, Swiss Alps) it is demonstrated that the developed time-lapse petrophysical joint inversion provides physically plausible solutions, in particular improved estimates for the volumetric fractions of ice and rock. The field application is evaluated with independent validation data including thaw depths derived from borehole temperature measurements and shows generally good agreement. As opposed to the conventional petrophysical joint inversion, its time-lapse extension succeeds in providing reasonable estimates of permafrost degradation at the Schilthorn monitoring site without a priori constraints on the porosity model, amounting to an approximate relative loss of 40% of ground ice in the upper 15m below the surface in the time period between 2008 and 2017.
-
Influence of prior data on the optimization of borehole locations for the estimation of geothermal reservoir parameters
2020 |
B.Sc. Applied Geosciences
Note: This thesis was supervised together with Johanna Fink and conducted during Florian's time as substitute professor of the Institute for Applied Geophysics and Geothermal Energy (now Computational Geoscience, Geothermics and Reservoir Geophysics).