Nina Kukowski
Research Subjects
Nina Kukowski has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
- Valentin Kasburg
Institute of Geosciences Friedrich Schiller University Jena Jena Germany
7 shared publications Andreas Goepel
5 shared publicationsCornelius Schwarze
3 shared publications- Cornelius Schwarze
Friedrich Schiller Univ Jena, Inst Geosci, Burgweg 11, D-07749 Jena, Germany
3 shared publications - Michael Warsitzka
Institute of Geophysics of the Czech Academy of Sciences, Boční II/1401, 14131 Prague, Czechia
3 shared publications Thomas Jahr
3 shared publications
Latest Publications
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Physics-Guided Neural Network Parameter Estimation of Groundwater–Strain Coupling at Moxa Geodynamic Observatory
High-resolution laser strainmeter measurements in an underground gallery at Moxa Geodynamic Observatory (Thuringia, central Germany) provide detailed records of crustal deformation. Beyond Earth tides, deformation induced by pore-pressure fluctuations produces the largest signals. These observations reveal that the relationship between groundwater transport and crustal strain is temporally fluctuating, highlighting the need for a quantitative approach to systematically characterize potential coupling.Here, we present a physics-guided, data-driven approach for estimating effective groundwater–strain coupling from multivariate time series of groundwater levels and nanometer-scale strain measurements, based on linear Biot poro-elasticity. The approach incorporates physically guided Biot neurons into an autoregressive neural network architecture; these neurons model horizontal poro-elastic responses of fractured rock driven by groundwater variations. It dynamically adjusts to temporal changes in groundwater levels and the resulting pore-pressure-induced strain. Using orientation-specific laser strainmeter measurements and spatially distributed groundwater levels from boreholes, we estimate Biot coupling in two horizontal directions (North–South and East–West) and derive effective coupling parameters from over a decade of observatory records.Our results provide insights into the dynamic hydro-mechanical behaviour of the shallow crust and highlight the potential of physics-guided neural architectures to support the interpretation of high-resolution deformation and stress–strain responses in geomechanical studies.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessPhysics-Guided Neural Network Parameter Estimation of Groundwater–Strain Coupling at Moxa Geodynamic Observatory
High-resolution laser strainmeter measurements in an underground gallery at Moxa Geodynamic Observatory (Thuringia, central Germany) provide detailed records of crustal deformation. Beyond Earth tides, deformation induced by pore-pressure fluctuations produces the largest signals. These observations reveal that the relationship between groundwater transport and crustal strain is temporally fluctuating, highlighting the need for a quantitative approach to systematically characterize potential coupling.Here, we present a physics-guided, data-driven approach for estimating effective groundwater–strain coupling from multivariate time series of groundwater levels and nanometer-scale strain measurements, based on linear Biot poro-elasticity. The approach incorporates physically guided Biot neurons into an autoregressive neural network architecture; these neurons model horizontal poro-elastic responses of fractured rock driven by groundwater variations. It dynamically adjusts to temporal changes in groundwater levels and the resulting pore-pressure-induced strain. Using orientation-specific laser strainmeter measurements and spatially distributed groundwater levels from boreholes, we estimate Biot coupling in two horizontal directions (North–South and East–West) and derive effective coupling parameters from over a decade of observatory records.Our results provide insights into the dynamic hydro-mechanical behaviour of the shallow crust and highlight the potential of physics-guided neural architectures to support the interpretation of high-resolution deformation and stress–strain responses in geomechanical studies.
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Variable detachment strength along the Peruvian margin estimated from Critical Taper Analysis
Along the Peruvian margin, the subduction of the Nazca Plate beneath the South American Plate shows significant along-strike variability, including changes in slab dip and the subduction of major bathymetric features such as the Nazca Ridge and several fault zones. Seismic behavior along the Peruvian margin is likewise highly variable, ranging from frequent large megathrust earthquakes in southern Peru to comparably low seismicity in the north, where tsunami earthquakes have nevertheless caused significant historical damage.To investigate the mechanical strength of the Peruvian forearc, we perform an areal critical taper analysis based on gridded surface slope and slab dip data. The results reveal significant spatial variations in detachment strength along the margin. South of the Nazca Ridge (~15°S), the forearc is characterized by a relatively strong detachment. The central segment (15°S–10°S) shows moderate to weak detachment strength, with particularly weak conditions near the trench. In northern Peru (10°S–4°S), the detachment is generally weak across the entire forearc. Overall, detachment strength tends to decrease toward the trench, except in the region affected by the subducting Nazca Ridge, where an increase in strength is observed.The long-time-scale spatial variability in detachment strength correlates fairly well with interseismic coupling patterns derived from short-time-scale geodetic observations, with locked portions of the subduction interface generally characterized by higher detachment strength. In our study, we address the question of whether and how short-term seismic behavior is controlled by the long-term mechanical properties of the Peruvian forearc.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessVariable detachment strength along the Peruvian margin estimated from Critical Taper Analysis
Along the Peruvian margin, the subduction of the Nazca Plate beneath the South American Plate shows significant along-strike variability, including changes in slab dip and the subduction of major bathymetric features such as the Nazca Ridge and several fault zones. Seismic behavior along the Peruvian margin is likewise highly variable, ranging from frequent large megathrust earthquakes in southern Peru to comparably low seismicity in the north, where tsunami earthquakes have nevertheless caused significant historical damage.To investigate the mechanical strength of the Peruvian forearc, we perform an areal critical taper analysis based on gridded surface slope and slab dip data. The results reveal significant spatial variations in detachment strength along the margin. South of the Nazca Ridge (~15°S), the forearc is characterized by a relatively strong detachment. The central segment (15°S–10°S) shows moderate to weak detachment strength, with particularly weak conditions near the trench. In northern Peru (10°S–4°S), the detachment is generally weak across the entire forearc. Overall, detachment strength tends to decrease toward the trench, except in the region affected by the subducting Nazca Ridge, where an increase in strength is observed.The long-time-scale spatial variability in detachment strength correlates fairly well with interseismic coupling patterns derived from short-time-scale geodetic observations, with locked portions of the subduction interface generally characterized by higher detachment strength. In our study, we address the question of whether and how short-term seismic behavior is controlled by the long-term mechanical properties of the Peruvian forearc.
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Influence of anisotropic rock properties on the regional temperature distribution in the Thuringian Basin, central Germany: A numerical simulation study
Temperature distribution in the geological subsurface of sedimentary basins is controlled by conductive and advective heat transport processes. The efficiency of these mechanisms is largely governed by the petrophysical properties of the rocks present in a sedimentary basin, with thermal conductivity being of primary importance for conductive heat transfer and permeability for advective heat transport. Sedimentary rocks exhibit a stratification-related anisotropy of both thermal conductivity and permeability, which plays a crucial role in the numerical assessment of thermal conditions in sedimentary basins.In this study, numerical 3D simulations are carried out using the Thuringian Basin in central Germany as a case study to investigate the influence of anisotropic thermal conductivity and permeability of its Permo-Triassic sedimentary infill on the regional temperature distribution. The modeled results are compared with available temperature measurements in drill holes to evaluate the predictive capability of the simulations. In addition, sensitivity analyses are performed to validate and contextualize the results by quantifying the significance of anisotropic parameters relative to other controlling factors.The simulation results indicate that accounting for the directional dependence of thermal conductivity and permeability has a significant impact on the regional temperature distribution. In particular, permeability anisotropy exerts a strong control on both the spatial position and the lateral extent of thermal anomalies. Models employing isotropic parameters may significantly underestimate the influence of stratification-induced anisotropy on the subsurface temperature distribution.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessInfluence of anisotropic rock properties on the regional temperature distribution in the Thuringian Basin, central Germany: A numerical simulation study
Temperature distribution in the geological subsurface of sedimentary basins is controlled by conductive and advective heat transport processes. The efficiency of these mechanisms is largely governed by the petrophysical properties of the rocks present in a sedimentary basin, with thermal conductivity being of primary importance for conductive heat transfer and permeability for advective heat transport. Sedimentary rocks exhibit a stratification-related anisotropy of both thermal conductivity and permeability, which plays a crucial role in the numerical assessment of thermal conditions in sedimentary basins.In this study, numerical 3D simulations are carried out using the Thuringian Basin in central Germany as a case study to investigate the influence of anisotropic thermal conductivity and permeability of its Permo-Triassic sedimentary infill on the regional temperature distribution. The modeled results are compared with available temperature measurements in drill holes to evaluate the predictive capability of the simulations. In addition, sensitivity analyses are performed to validate and contextualize the results by quantifying the significance of anisotropic parameters relative to other controlling factors.The simulation results indicate that accounting for the directional dependence of thermal conductivity and permeability has a significant impact on the regional temperature distribution. In particular, permeability anisotropy exerts a strong control on both the spatial position and the lateral extent of thermal anomalies. Models employing isotropic parameters may significantly underestimate the influence of stratification-induced anisotropy on the subsurface temperature distribution.
- Open Access
- 11 February 2026
- Published by Copernicus GmbH in Scientific Drilling
- Vol. 35 (1)
Follow the CO 2 – drilling into an actively degassing intraplate volcano underlain by a silicate–carbonatite intrusion
The Eifel region of Germany hosts hundreds of distributed volcanoes of Quaternary age in an intracontinental setting. This includes many maar volcanoes, for which the Eifel is the type locality. Laacher See volcano in the eastern part of the region stands out as a sizable (erupted volume of 6.3 km3 dense rock equivalent or a volcanic explosivity index (VEI) of 6) and dormant but actively deforming and degassing system. Plutonic ejecta clasts in pyroclastic deposits of the Laacher See volcano provide evidence that it is underlain by one of the youngest silicate–carbonatite subvolcanic intrusive complexes worldwide. It has long been appreciated that the Laacher See region has potential to significantly enhance our knowledge on distributed volcanic fields and their specific hazards resulting from high CO2 fluxes from the mantle to the surface, causing active deformation, as well as diffuse and sometimes punctuated explosive degassing. This is largely due to the Eifel boasting an extensive record of past research, easy access, and excellent infrastructure that uniquely permits implementation of cutting-edge scientific methods. Recently, three workshops were held to sharpen scientific questions of global scope and relevance to study this type of distributed volcanism. Workshop participants discussed opportunities and challenges associated with drilling in the Laacher See region, identified promising sites, and explored the potential of novel drilling techniques. The clear conclusion of these workshops is that Laacher See would be an ideal test bed to evaluate the physical and chemical properties of a shallow (∼ 4–6 km depth at its top, thus making it accessible to drilling) silicate–carbonatite intrusive complex formed by volatile-rich melts and associated with active degassing. Only drilling can provide answers to key problems related to the geodynamics, geohazards, and resource potential of such magma systems. Questions include how and at what rates translithospheric transport of magma and fluids occurs in continental intraplate settings. It is also puzzling why shallow fluids above a residual magma system after an eruption that occurred only 13 000 years ago are seemingly cold, although the system still appears to be highly dynamic based on ongoing CO2 degassing, seismic activity, and exceptionally high uplift rates on spatial scales of hundreds of kilometres. Critically, drilling enables linking deep and shallow observables related to melt and fluid migration and provides access to samples of CO2-rich fluids and their host rocks at depth. Such samples are tangible evidence required to properly balance CO2 fluxes from degassing magma with CO2 sequestration in carbonatites or fluid-precipitated carbonates. Core samples from a maar structure proximally to the Laacher See volcano can establish an unprecedented geological record of precursor events prior to its cataclysmic eruption. Importantly, drilling also permits improved geophysical monitoring from instrumented wells that can reveal high-resolution details on maar diatreme architecture and deflation-related faulting resulting from the evacuation of the Laacher See magma reservoir. Lastly, silicate–carbonatite intrusions are globally recognised as major hosts for critical metal deposits, which, at Laacher See, could be topics of investigation in the making. Ultimately, this project can provide fundamental insights into processes of fluid-mediated element transport and sequestration not achievable in inactive carbonatites. Overall, these goals are best achieved in two phases that encompass drilling (1) four holes (300–2000 m) to enable detailed studies of the pre-eruptive evolution of the Laacher See volcano, its subvolcanic structures, and ongoing fluid or magma transport and (2) a subsequent deep hole (3000–4000 m) to penetrate and core a syenitic–carbonatitic intrusive carapace and its hydrothermal aureole.
- 11 February 2026
- Published by Copernicus GmbH in Scientific Drilling
- Vol. 35 (1)
Open AccessFollow the CO 2 – drilling into an actively degassing intraplate volcano underlain by a silicate–carbonatite intrusion
The Eifel region of Germany hosts hundreds of distributed volcanoes of Quaternary age in an intracontinental setting. This includes many maar volcanoes, for which the Eifel is the type locality. Laacher See volcano in the eastern part of the region stands out as a sizable (erupted volume of 6.3 km3 dense rock equivalent or a volcanic explosivity index (VEI) of 6) and dormant but actively deforming and degassing system. Plutonic ejecta clasts in pyroclastic deposits of the Laacher See volcano provide evidence that it is underlain by one of the youngest silicate–carbonatite subvolcanic intrusive complexes worldwide. It has long been appreciated that the Laacher See region has potential to significantly enhance our knowledge on distributed volcanic fields and their specific hazards resulting from high CO2 fluxes from the mantle to the surface, causing active deformation, as well as diffuse and sometimes punctuated explosive degassing. This is largely due to the Eifel boasting an extensive record of past research, easy access, and excellent infrastructure that uniquely permits implementation of cutting-edge scientific methods. Recently, three workshops were held to sharpen scientific questions of global scope and relevance to study this type of distributed volcanism. Workshop participants discussed opportunities and challenges associated with drilling in the Laacher See region, identified promising sites, and explored the potential of novel drilling techniques. The clear conclusion of these workshops is that Laacher See would be an ideal test bed to evaluate the physical and chemical properties of a shallow (∼ 4–6 km depth at its top, thus making it accessible to drilling) silicate–carbonatite intrusive complex formed by volatile-rich melts and associated with active degassing. Only drilling can provide answers to key problems related to the geodynamics, geohazards, and resource potential of such magma systems. Questions include how and at what rates translithospheric transport of magma and fluids occurs in continental intraplate settings. It is also puzzling why shallow fluids above a residual magma system after an eruption that occurred only 13 000 years ago are seemingly cold, although the system still appears to be highly dynamic based on ongoing CO2 degassing, seismic activity, and exceptionally high uplift rates on spatial scales of hundreds of kilometres. Critically, drilling enables linking deep and shallow observables related to melt and fluid migration and provides access to samples of CO2-rich fluids and their host rocks at depth. Such samples are tangible evidence required to properly balance CO2 fluxes from degassing magma with CO2 sequestration in carbonatites or fluid-precipitated carbonates. Core samples from a maar structure proximally to the Laacher See volcano can establish an unprecedented geological record of precursor events prior to its cataclysmic eruption. Importantly, drilling also permits improved geophysical monitoring from instrumented wells that can reveal high-resolution details on maar diatreme architecture and deflation-related faulting resulting from the evacuation of the Laacher See magma reservoir. Lastly, silicate–carbonatite intrusions are globally recognised as major hosts for critical metal deposits, which, at Laacher See, could be topics of investigation in the making. Ultimately, this project can provide fundamental insights into processes of fluid-mediated element transport and sequestration not achievable in inactive carbonatites. Overall, these goals are best achieved in two phases that encompass drilling (1) four holes (300–2000 m) to enable detailed studies of the pre-eruptive evolution of the Laacher See volcano, its subvolcanic structures, and ongoing fluid or magma transport and (2) a subsequent deep hole (3000–4000 m) to penetrate and core a syenitic–carbonatitic intrusive carapace and its hydrothermal aureole.
- Open Access
- 18 March 2025
- Published by Copernicus GmbH
Thermal Conductivity Measurements of in Resin-Embedded Cuttings Using the Optical Scanning Method
The transport of heat from the Earth’s interior to its surface is the maindriving force for many geological processes, like e.g. lithosphere dynamics andmantle convection. To understand these processes, it is essential to capturethe spatially and temporally dependent temperature field of the Earth’s crust.The structure of this temperature field is largely determined by the thermalproperties of the rocks. One of the key thermophysical properties is thermalconductivity, which plays a critical role in various scientific and economic fields,such as the study of sedimentary basins and geothermal potentials.To characterize the thermal properties of various lithologies, measurementson core samples are of ultimate interest. However, as coring is highly expen-sive, drill cores are not often taken during drilling campaigns. Nonetheless, drillcuttings would be available from nearly all drill holes. The drill cuttings canbe ground and compressed to form measuring tablets for thermal conductivitytests, which changes the rock structure and thus also the thermal properties. Analternative could be to embed drill cuttings in materials of known thermal prop-erties like resin to prepare samples ready to be used with the widely availableThermal Conductivity Scanner (TCS).In this study, we present an innovative method for determining the thermalconductivity of drill cuttings using TCS by preparing various samples drill cut-tings embedded in epoxy resin. We investigate how grain size of the encasedcuttings affects the estimation of thermal conductivity with the TCS. To testthis novel approach, we use artificially produced drill cuttings, mini-cores em-bedded in epoxy resin, and undisturbed reference samples. Our investigationssuggest that grain size and thermal edge effects resulting from embedding thesamples in resin can be corrected with sufficient precision to enable accurateconclusions about the thermal conductivity of the undisturbed rock.
- 18 March 2025
- Published by Copernicus GmbH
Open AccessThermal Conductivity Measurements of in Resin-Embedded Cuttings Using the Optical Scanning Method
The transport of heat from the Earth’s interior to its surface is the maindriving force for many geological processes, like e.g. lithosphere dynamics andmantle convection. To understand these processes, it is essential to capturethe spatially and temporally dependent temperature field of the Earth’s crust.The structure of this temperature field is largely determined by the thermalproperties of the rocks. One of the key thermophysical properties is thermalconductivity, which plays a critical role in various scientific and economic fields,such as the study of sedimentary basins and geothermal potentials.To characterize the thermal properties of various lithologies, measurementson core samples are of ultimate interest. However, as coring is highly expen-sive, drill cores are not often taken during drilling campaigns. Nonetheless, drillcuttings would be available from nearly all drill holes. The drill cuttings canbe ground and compressed to form measuring tablets for thermal conductivitytests, which changes the rock structure and thus also the thermal properties. Analternative could be to embed drill cuttings in materials of known thermal prop-erties like resin to prepare samples ready to be used with the widely availableThermal Conductivity Scanner (TCS).In this study, we present an innovative method for determining the thermalconductivity of drill cuttings using TCS by preparing various samples drill cut-tings embedded in epoxy resin. We investigate how grain size of the encasedcuttings affects the estimation of thermal conductivity with the TCS. To testthis novel approach, we use artificially produced drill cuttings, mini-cores em-bedded in epoxy resin, and undisturbed reference samples. Our investigationssuggest that grain size and thermal edge effects resulting from embedding thesamples in resin can be corrected with sufficient precision to enable accurateconclusions about the thermal conductivity of the undisturbed rock.
- Open Access
- 18 March 2025
- Published by Copernicus GmbH
Partitioning of deformation along the Andean margin: insights from elastic block modelling
Along the Andean margin, oblique subduction of the Nazca Plate is accommodated by slip on the subduction interface and deformation of the overriding South American Plate. Active plate boundary deformation, in particular due to strain partitioning, is analyzed using elastic block modelling constrained by compiled GPS velocities to estimate plate motions, fault slip rates, and spatially variable interplate coupling on the Nazca-South American subduction interface. In the block modelling approach, interseismic GPS velocities are assumed to be the sum of rigid block rotation and elastic strain accumulation on block-bounding faults. Therefore, the western South American margin is divided into smaller blocks, primarily based on active faults traces. The block model geometry is adjusted to minimize the misfit between observed and modeled velocities.The preferred model shows strain partitioning of varying degrees along the Andean margin. In the North, the margin-parallel component of convergence is partially accommodated by right-lateral slip on a strike-slip system that extends from the Gulf of Guayaquil off southern Ecuador to western Venezuela. This results in the northeastward motion of the North Andean Block with respect to stable South America. In Peru, the model confirms the existence of the southeastward moving Inca Sliver, that is bounded by the trench and the Subandean fold-and-thrust belt. Along the central Chilean margin, oblique convergence is partially accommodated by minor right-lateral slip along the Subandean thrust fault. In southern Chile, right-lateral transpression along the intra-arc Liquiñe-Ofqui Fault Zone results in the northward translation of the Chiloé Sliver.The separation of the North Andean Block, the Peruvian Inca Sliver, and the Chilean Andean orogen is related to the curvature of the Andean margin and the associated changes in the sense of convergence obliquity. The differing directions of movement of these blocks result in extension in the Gulf of Guayaquil, where the Andean margin is seaward convex. In contrast, on the Altiplano, at the concave bend of western South America, the rotational velocities of the Peruvian and Chilean blocks are converging into a similar direction.The spatial distribution of interplate coupling as estimated by our block modelling shows that the Andean margin is segmented into strongly and weakly coupled zones. Epicenters of major thrust earthquakes correlate fairly well with areas of strong interplate coupling.
- 18 March 2025
- Published by Copernicus GmbH
Open AccessPartitioning of deformation along the Andean margin: insights from elastic block modelling
Along the Andean margin, oblique subduction of the Nazca Plate is accommodated by slip on the subduction interface and deformation of the overriding South American Plate. Active plate boundary deformation, in particular due to strain partitioning, is analyzed using elastic block modelling constrained by compiled GPS velocities to estimate plate motions, fault slip rates, and spatially variable interplate coupling on the Nazca-South American subduction interface. In the block modelling approach, interseismic GPS velocities are assumed to be the sum of rigid block rotation and elastic strain accumulation on block-bounding faults. Therefore, the western South American margin is divided into smaller blocks, primarily based on active faults traces. The block model geometry is adjusted to minimize the misfit between observed and modeled velocities.The preferred model shows strain partitioning of varying degrees along the Andean margin. In the North, the margin-parallel component of convergence is partially accommodated by right-lateral slip on a strike-slip system that extends from the Gulf of Guayaquil off southern Ecuador to western Venezuela. This results in the northeastward motion of the North Andean Block with respect to stable South America. In Peru, the model confirms the existence of the southeastward moving Inca Sliver, that is bounded by the trench and the Subandean fold-and-thrust belt. Along the central Chilean margin, oblique convergence is partially accommodated by minor right-lateral slip along the Subandean thrust fault. In southern Chile, right-lateral transpression along the intra-arc Liquiñe-Ofqui Fault Zone results in the northward translation of the Chiloé Sliver.The separation of the North Andean Block, the Peruvian Inca Sliver, and the Chilean Andean orogen is related to the curvature of the Andean margin and the associated changes in the sense of convergence obliquity. The differing directions of movement of these blocks result in extension in the Gulf of Guayaquil, where the Andean margin is seaward convex. In contrast, on the Altiplano, at the concave bend of western South America, the rotational velocities of the Peruvian and Chilean blocks are converging into a similar direction.The spatial distribution of interplate coupling as estimated by our block modelling shows that the Andean margin is segmented into strongly and weakly coupled zones. Epicenters of major thrust earthquakes correlate fairly well with areas of strong interplate coupling.
- Open Access
- 18 March 2025
- Published by Copernicus GmbH
Automating Seismic Event Discrimination: A Comparative Study of Convolutional Neural Networks and Vision Transformers
To date, the discrimination of seismic events recorded in seismic networks is often performed manually by experts, classifying events into categories such as earthquakes, quarry blasts, or mining events. While recent studies have shown that deep learning algorithms, particularly Convolutional Neural Networks (CNNs), can efficiently and accurately distinguish between different types of seismic events, their application for automated seismic event discrimination remains limited. This limitation arises from several factors, including the absence of globally applicable models that maintain high precision for local seismic networks, the scarcity of data required for fine-tuning Deep Learning (DL) models, and the lack of interpretability in the decision-making processes of these black-box models.In this contribution, we explore the use of Vision Transformers (ViTs) as a novel approach for automating seismic event discrimination. To assess their potential for accuracy and explainability, we applied CNNs and ViTs to classify seismic events such as earthquakes, quarry blasts, and mining events. For this purpose, we pretrained the models on openly available seismic event data from Utah and Northern California and then fine-tuned and tested them on data from the Seismic Network of the Ruhr-University Bochum (RuhrNet) and the Thuringian Seismic Network (TSN).Our findings reveal that ViTs can analyze the entire spectrogram of a seismic event in a coherent manner, offering superior generalizability in pattern recognition compared to CNNs. In addition to achieving high discrimination accuracy, the attention weights of ViTs provide insights into the models’ decision-making process, offering a transparent and interpretable explanation of the underlying mechanisms driving its classifications.
- 18 March 2025
- Published by Copernicus GmbH
Open AccessAutomating Seismic Event Discrimination: A Comparative Study of Convolutional Neural Networks and Vision Transformers
To date, the discrimination of seismic events recorded in seismic networks is often performed manually by experts, classifying events into categories such as earthquakes, quarry blasts, or mining events. While recent studies have shown that deep learning algorithms, particularly Convolutional Neural Networks (CNNs), can efficiently and accurately distinguish between different types of seismic events, their application for automated seismic event discrimination remains limited. This limitation arises from several factors, including the absence of globally applicable models that maintain high precision for local seismic networks, the scarcity of data required for fine-tuning Deep Learning (DL) models, and the lack of interpretability in the decision-making processes of these black-box models.In this contribution, we explore the use of Vision Transformers (ViTs) as a novel approach for automating seismic event discrimination. To assess their potential for accuracy and explainability, we applied CNNs and ViTs to classify seismic events such as earthquakes, quarry blasts, and mining events. For this purpose, we pretrained the models on openly available seismic event data from Utah and Northern California and then fine-tuned and tested them on data from the Seismic Network of the Ruhr-University Bochum (RuhrNet) and the Thuringian Seismic Network (TSN).Our findings reveal that ViTs can analyze the entire spectrogram of a seismic event in a coherent manner, offering superior generalizability in pattern recognition compared to CNNs. In addition to achieving high discrimination accuracy, the attention weights of ViTs provide insights into the models’ decision-making process, offering a transparent and interpretable explanation of the underlying mechanisms driving its classifications.
- 27 December 2023
- Published by Seismological Society of America (SSA) in Bulletin of the Seismological Society of America
- Vol. 114 (2) , 842-856
Cross-Regional Seismic Event Discrimination via Convolutional Neural Networks: Exploring Fine-Tuning and Ensemble Averaging
The gradual densification of seismic networks has facilitated the acquisition of large amounts of data. However, alongside natural tectonic earthquakes, seismic networks also record anthropogenic events such as quarry blasts or other induced events. Identifying and distinguishing these events from natural earthquakes requires experienced interpreters to ensure that seismological studies of natural phenomena are not compromised by anthropogenic events. Advanced artificial intelligence methods have already been deployed to tackle this problem. One of the applications includes Convolutional Neural Networks (CNN) to discriminate different kinds of events, such as natural earthquakes and quarry blasts. In this study, we investigate the effects of ensemble averaging and fine-tuning on seismic event discrimination accuracy to estimate the potential of these methods. We compare discrimination accuracy of two different CNN model architectures across three datasets. This was done with the best models from an ensemble of each model architecture, as well as with ensemble averaging and fine-tuning methods. Soft voting was used for the CNN ensemble predictions. For the transfer learning approach, the models were pretrained with data from two of the datasets (nontarget regions) and fine-tuned with data from the third one (target region). The results show that ensemble averaging and fine-tuning of CNN models leads to better generalization of the model predictions. For the region with the lowest numbers of one event type, the combination of ensemble averaging and fine-tuning led to an increase in discrimination accuracy of up to 4% at station level and up to 10% at event level. We also tested the impact of the amount of training data on the fine-tuning method, showing, that to create a global model, the selection of comprehensive training data is needed.
- 27 December 2023
- Published by Seismological Society of America (SSA) in Bulletin of the Seismological Society of America
- Vol. 114 (2) , 842-856
Cross-Regional Seismic Event Discrimination via Convolutional Neural Networks: Exploring Fine-Tuning and Ensemble Averaging
The gradual densification of seismic networks has facilitated the acquisition of large amounts of data. However, alongside natural tectonic earthquakes, seismic networks also record anthropogenic events such as quarry blasts or other induced events. Identifying and distinguishing these events from natural earthquakes requires experienced interpreters to ensure that seismological studies of natural phenomena are not compromised by anthropogenic events. Advanced artificial intelligence methods have already been deployed to tackle this problem. One of the applications includes Convolutional Neural Networks (CNN) to discriminate different kinds of events, such as natural earthquakes and quarry blasts. In this study, we investigate the effects of ensemble averaging and fine-tuning on seismic event discrimination accuracy to estimate the potential of these methods. We compare discrimination accuracy of two different CNN model architectures across three datasets. This was done with the best models from an ensemble of each model architecture, as well as with ensemble averaging and fine-tuning methods. Soft voting was used for the CNN ensemble predictions. For the transfer learning approach, the models were pretrained with data from two of the datasets (nontarget regions) and fine-tuned with data from the third one (target region). The results show that ensemble averaging and fine-tuning of CNN models leads to better generalization of the model predictions. For the region with the lowest numbers of one event type, the combination of ensemble averaging and fine-tuning led to an increase in discrimination accuracy of up to 4% at station level and up to 10% at event level. We also tested the impact of the amount of training data on the fine-tuning method, showing, that to create a global model, the selection of comprehensive training data is needed.
- Open Access
- 15 May 2023
- Published by Copernicus GmbH
Patterns and cyclicity of Quaternary sedimentation above a subducting seamount at Rock Garden (central Hikurangi Margin, New Zealand)
Rock Garden (RG), located east off the North Island of New Zealand, is part of an accretionary ridge that is influenced by seamount subduction. Two ∼37m long sediment cores, drilled with the seafloor drill rig (MeBo200) from RG, provide a continuous sedimentary record of the period between 1.95-0.4 Ma. This period, the Early-Middle Pleistocene transition (EMPT), was marked by a progressive increase in the amplitude of climate oscillations and a shift of Milanković cycles from 41 ka towards a quasi-100 ka frequency in the absence of any significant change in orbital forcing. From the recovered core material of cores GeoB20824-4 and GeoB20846-1, we determined sediment physical properties, oxygen isotope (δ18O) values, and element concentrations based on X-ray fluorescence (XRF) measurements. The element ratios were used as proxies for sediment composition and as paleoenvironmental indices. In sediment physical properties, δ18O values, and geochemical properties, evidence for glacial and interglacial cycles and cyclicities of 405 ka, 100 ka, 41 ka were found. A shift of the cyclicity from 41 to 100 ka took place in sediment cores during (1.4-0.4 Ma). Numerical ages obtained from tephra layers included in the sedimentary record enabled to estimate sedimentation rates from both cores. Although both drill sites are only 1800 m away from each other, sedimentation rates of 2.15-2.96 cmka−1 (GeoB20824-4) and 5.49-6.77 cmka−1 (GeoB20846-1), respectively, differ by a factor of two. This may be the reason why two facies-units were identified in core GeoB20824-4, whereas sediments of core GeoB20846-1 all belonged to the same facies. A change of lithofacies in core GeoB20824-4 between Unit I and Unit II in ∼20 mbsf at 1.5-1.4 Ma marks the initiation of the EMPT.
- 15 May 2023
- Published by Copernicus GmbH
Open AccessPatterns and cyclicity of Quaternary sedimentation above a subducting seamount at Rock Garden (central Hikurangi Margin, New Zealand)
Rock Garden (RG), located east off the North Island of New Zealand, is part of an accretionary ridge that is influenced by seamount subduction. Two ∼37m long sediment cores, drilled with the seafloor drill rig (MeBo200) from RG, provide a continuous sedimentary record of the period between 1.95-0.4 Ma. This period, the Early-Middle Pleistocene transition (EMPT), was marked by a progressive increase in the amplitude of climate oscillations and a shift of Milanković cycles from 41 ka towards a quasi-100 ka frequency in the absence of any significant change in orbital forcing. From the recovered core material of cores GeoB20824-4 and GeoB20846-1, we determined sediment physical properties, oxygen isotope (δ18O) values, and element concentrations based on X-ray fluorescence (XRF) measurements. The element ratios were used as proxies for sediment composition and as paleoenvironmental indices. In sediment physical properties, δ18O values, and geochemical properties, evidence for glacial and interglacial cycles and cyclicities of 405 ka, 100 ka, 41 ka were found. A shift of the cyclicity from 41 to 100 ka took place in sediment cores during (1.4-0.4 Ma). Numerical ages obtained from tephra layers included in the sedimentary record enabled to estimate sedimentation rates from both cores. Although both drill sites are only 1800 m away from each other, sedimentation rates of 2.15-2.96 cmka−1 (GeoB20824-4) and 5.49-6.77 cmka−1 (GeoB20846-1), respectively, differ by a factor of two. This may be the reason why two facies-units were identified in core GeoB20824-4, whereas sediments of core GeoB20846-1 all belonged to the same facies. A change of lithofacies in core GeoB20824-4 between Unit I and Unit II in ∼20 mbsf at 1.5-1.4 Ma marks the initiation of the EMPT.
- Open Access
- 15 May 2023
- Published by Copernicus GmbH
Influence of Environmental Parameters on Highly Sensitive Instruments at Moxa Geodynamic Observatory
Strainmeters measure the change in length between two known fixed points and are used primarily to identify and estimate tectonic strain. In addition to tectonic strain, these instruments also record changes in strain caused by other phenomena such as Earth tides and fluctuations of meteorological signals like changes in groundwater levels caused by precipitation, which may be much larger than tectonic strain and thus mask its signals. To avoid meteorological influences as far as possible, horizontal strainmeters often are maintained in galleries such that tectonically induced strain signals are not affected by other sources of noise.At Moxa Geodynamic Observatory, located in central Germany, two laser strainmeters with a base length of 26 m each are maintained in galleries oriented north-south and east-west. Their resolution is in the nano-scale and sampling rate is 0.1 Hz. Mountain overburden in the gallery is comparatively low at approx. 30 m and in addition, the hydrogeological situation of the subsurface surrounding the observatory, is very heterogeneous. Therefore, amplitudes of meteorological phenomena are still quite high. In order to correct meteorological influences in the recorded strain time-series, they first need to be better understood. For doing so, long time-series - a decade at least – are needed. As the laser strainmeters continuously record since summer 2011, these are now available.We present the results of weighting various meteorological parameters on the strainmeter recordings by training Long Short Term Memory Networks and perturbing input parameters for the test data. In this way, the contribution of each parameter to the meteorologically induced strain signals can be estimated. This knowledge is subsequently used to eliminate meteorological influences from the time-series recordings of strain.
- 15 May 2023
- Published by Copernicus GmbH
Open AccessInfluence of Environmental Parameters on Highly Sensitive Instruments at Moxa Geodynamic Observatory
Strainmeters measure the change in length between two known fixed points and are used primarily to identify and estimate tectonic strain. In addition to tectonic strain, these instruments also record changes in strain caused by other phenomena such as Earth tides and fluctuations of meteorological signals like changes in groundwater levels caused by precipitation, which may be much larger than tectonic strain and thus mask its signals. To avoid meteorological influences as far as possible, horizontal strainmeters often are maintained in galleries such that tectonically induced strain signals are not affected by other sources of noise.At Moxa Geodynamic Observatory, located in central Germany, two laser strainmeters with a base length of 26 m each are maintained in galleries oriented north-south and east-west. Their resolution is in the nano-scale and sampling rate is 0.1 Hz. Mountain overburden in the gallery is comparatively low at approx. 30 m and in addition, the hydrogeological situation of the subsurface surrounding the observatory, is very heterogeneous. Therefore, amplitudes of meteorological phenomena are still quite high. In order to correct meteorological influences in the recorded strain time-series, they first need to be better understood. For doing so, long time-series - a decade at least – are needed. As the laser strainmeters continuously record since summer 2011, these are now available.We present the results of weighting various meteorological parameters on the strainmeter recordings by training Long Short Term Memory Networks and perturbing input parameters for the test data. In this way, the contribution of each parameter to the meteorologically induced strain signals can be estimated. This knowledge is subsequently used to eliminate meteorological influences from the time-series recordings of strain.