Federica Lanza

Last Name

Lanza

First Name

Federica

ORCID

0000-0002-8168-6766

Organisational unit

02818 - Schweiz. Erdbebendienst (SED) / Swiss Seismological Service (SED)

Show all metadata

Search Results

Publications 1 - 10 of 12

The bound growth of induced earthquakes could de-risk hydraulic fracturing
Schultz, Ryan; Lanza, Federica; Dyer, Ben; et al. (2025)
Communications Earth & Environment
The world’s energy supply depends critically on hydraulic fracturing (HF) to access otherwise uneconomical resources. Unfortunately, HF also has the potential to induce larger earthquakes – with some projects being prematurely terminated because of perceived earthquake risks. To de-risk HF, we use a suite of statistical tests to discern if some physical process has restricted the growth of earthquake magnitudes. We show that all stage stimulations at both UK PNR-1z and Helsinki St1 indicate bound fracture growth, implying a more controllable operation. Contrastingly, stimulations at Utah FORGE and UK PNR-2 sequentially transitioned into unbound fault reactivation. The problematic stages (that ultimately led to the termination of PNR-2) are clearly distinguishable. We postulate that our research can discriminate fracture stimulation from fault reactivation, contributing to the de-risking of HF operations worldwide. Our statistical tests provide a framework for model falsification, which can guide physical insights into the bounding processes.
From Labquakes to Megathrusts: Scaling Deep Learning Based Pickers Over 15 Orders of Magnitude
Shi, Peidong; Meier, Men‐Andrin; Villiger, Linus; et al. (2024)
Journal of Geophysical Research: Machine Learning and Computation
The application of machine learning techniques in seismology has greatly advanced seismological analysis, especially for earthquake detection and seismic phase picking. However, machine learning approaches still face challenges in generalizing to data sets that differ from their original training setting. Previous studies focused on retraining or transfer-learning models for these scenarios, but require high-quality labeled data sets. This paper demonstrates a new approach for augmenting already trained models without the need for additional training data. We propose four strategies—rescaling, model aggregation, shifting, and filtering—to enhance the performance of pre-trained models on out-of-distribution data sets. We further devise various methodologies to ensemble the individual predictions from these strategies to obtain a final unified prediction result featuring prediction robustness and detection sensitivity. We develop an open-source Python module quakephase that implements these methods and can flexibly process input continuous seismic data of any sampling rate. With quakephase and pre-trained ML models from SeisBench, we perform systematic benchmark tests on data recorded by different types of instruments, ranging from acoustic emission sensors to distributed acoustic sensing, and collected at different scales, spanning from laboratory acoustic emission events to major tectonic earthquakes. Our tests highlight that rescaling is essential for dealing with small-magnitude seismic events recorded at high sampling rates as well as larger magnitude events having long coda and remote events with long wave trains. Our results demonstrate that the proposed methods are effective in augmenting pre-trained models for out-of-distribution data sets, especially in scenarios with limited labeled data for transfer learning.
Insights into Extensional Tectonics from a Large Deep Learning Focal Mechanism Catalog
Meier, Men-Andrin; Lanza, Federica; Martínez-Garzón, Patricia; et al. (2025)
Bulletin of the Seismological Society of America
Earthquake sequences play out on geologic fault and fracture systems, which are usually underconstrained by data. Modern deep learning earthquake detection and characterization methods now allow us to compute high-sensitivity and high-resolution seismicity catalogs, with which we can image at least the seismogenic parts of fault and fracture systems with much more detail than had been possible previously. Here, we use a convolutional neural network classifier and the SKHASH algorithm to compute a catalog of 16,600 well-constrained focal mechanisms (FMs) for the exceptionally well-monitored 2016 M_w 6.0 Amatrice–M_w 5.9 Visso–M_w 6.5 Norcia earthquake sequence in Italy. The resulting catalog paints a detailed pic-ture of earthquake faulting kinematics in a fragmented extensional tectonic system. We observe that normal-faulting mechanisms dominate the seismic activity only over the depth range of 2–9 km. At shallower depths, for which the overburden may be too low for normal faults to be elastically loaded, strike-slip faulting is more common. The much-debated basal shear zone—an extensive about 2 km wide near-horizontal layer with distributed seismicity at 8–10 km depth—is characterized by much higher FM variability than the shallower parts of the crust, where the main normal faults host the largest earthquakes. In the north and south of the study region, the basal shear zone seismicity is sharply divided by the main normal faults, with predominantly normal faulting in the hanging wall, and predominantly strike-slip faulting in the footwall. The FMs from this study provide insight into deformation processes at the intersection of this basal shear zone and the major normal faults, which is where both the Amatrice and Norcia events nucleated.
Earthquakes in Switzerland and surrounding regions during 2019 and 2020
Diehl, Tobias; Cauzzi, Carlo Virgilio; Clinton, John Francis; et al. (2025)
Swiss Journal of Geosciences
This report summarizes the seismicity in Switzerland and surrounding regions in the years 2019 and 2020. In 2019 and 2020, the Swiss Seismological Service detected and located 1660 and 1407 earthquakes in the region under consideration, respectively. The strongest event in the analysed period was the ML 4.3 Elm/Steinibach earthquake, which occurred in the Glarus Alps in eastern Switzerland on October 25, 2020. Received felt reports suggest intensities up to degree V for this earthquake. Modelled and instrumentally measured ground motions, however, hint at intensities approaching degree VI-VII at the epicentre. Derived focal mechanisms and relative hypocentre relocations of fore- and aftershocks image a dextral WSW-ENE to W-E striking multi-segment strike-slip fault zone with a total length of about 3.5 km. Well-constrained focal depths of 1-2 km indicate that the fault zone likely locates in the uppermost part of the crystalline basement of the eastern Aar Massif. Another exceptional earthquake sequence occurred between Anz & egrave;re and Sanetschpass in the Rawil Depression in November 2019. Within 10 days, more than 300 earthquakes occurred in this cluster and 16 of those events reached ML magnitudes between 2.5 and 3.3. Focal mechanisms and relative hypocentre relocations derived for this sequence image the reactivation of a contractional stepover. The imaged stepover confirms the previously proposed segmented nature of the Rawil Fault Zone north of the Rh & ocirc;ne valley in SW Switzerland. The ML 4.2 Novel earthquake, which occurred in the Pr & eacute;alpes region south of Lake Geneva on May 28, 2019, provides additional evidence for the recently proposed domain of NE-SW oriented extensional to transtensional deformation along the Alpine Front in the transition zone between Central and Western Alps. Evidence for transtensional deformation along the SW edge of the Mont-Blanc Massif is provided by another remarkable earthquake cluster near the Grandes Jorasses Mountain in the border region between France and Italy. The transtensional deformation of the Hegau-Bodensee Graben in the northern foreland is revealed by a vigorous earthquake sequence on the Bodanr & uuml;ck Peninsula in southern Germany in 2019. Finally, evidence for unusually shallow seismicity in the domain of the Dent-Blanche nappe is provided by the ML 3.5 Arolla earthquake. In conclusion, the seismic activity during the period 2019-2020 is exceptional in terms of absolute numbers of earthquakes as well as number of events with ML >= 2.5.
Managing Induced Seismicity Risks From Enhanced Geothermal Systems: A Good Practice Guideline
Zhou, Wen; Lanza, Federica; Grigoratos, Iason; et al. (2024)
Reviews of Geophysics
Geothermal energy is a green source of power that could play an important role in climate-conscious energy portfolios; enhanced geothermal systems (EGS) have the potential to scale up exploitation of thermal resources. During hydraulic fracturing, fluids injected under high-pressure cause the rock mass to fail, stimulating fractures that improve fluid connectivity. However, this increase of pore fluid pressure can also reactivate pre-existing fault systems, potentially inducing earthquakes of significant size. Induced earthquakes are a significant concern for EGS operations. In some cases, ground shaking nuisance, building damages, or injuries have spurred the early termination of projects (e.g., Basel, Pohang). On the other hand, EGS operations at Soultz-sous-Forêts (France), Helsinki (Finland), Blue Mountain (Nevada, USA), and Utah FORGE (USA) have adequately managed induced earthquake risks. The success of an EGS operation depends on economical reservoir enhancements, while maintaining acceptable seismic risk levels. This requires state-of-the-art seismic risk management. This article reviews domains of seismology, earthquake engineering, risk management, and communication. We then synthesize "good practice" recommendations for evaluating, mitigating, and communicating the risk of induced seismicity. We advocate for a modular approach. Recommendations are provided for key technical aspects including (a) a seismic risk management framework, (b) seismic risk pre-screening, (c) comprehensive seismic hazard and risk evaluation, (d) traffic light protocol designs, (e) seismic monitoring implementation, and (f) step-by-step communication plans. Our recommendations adhere to regulatory best practices, to ensure their general applicability. Our guidelines provide a template for effective earthquake risk management and future research directions.
Toward a Digital Twin for Seismic Waves of Induced Events in a Geothermal Reservoir
Ermert, Laura; Lanza, Federica; Shi, Peidong; et al. (2025)
Bulletin of the Seismological Society of America
Monitoring induced seismicity is a key component of risk management for enhanced geothermal systems (EGSs). Induced seismicity presents particular monitoring requirements related to high event rates, small magnitudes, and difficulties of workflow testing due to the scarcity of manually labeled and ground-truth data sets. Digital replicas of geothermal reservoirs could help address these requirements. Here, we present a seismic wave propagation digital replica inspired by the Utah Frontier Observatory for Research in Geothermal Energy (FORGE). We simulate over 20,000 induced event waveforms with a spectral-element solver and source–receiver reciprocity, assemble them by origin time, and superimpose either low-amplitude Gaussian noise or site-specific correlated noise. The resulting continuous synthetic waveforms are used to test a real-time and a postprocessing monitoring workflow based on SeisComP and MALMI, respectively. Both workflows per-form well at retrieving input frequency–magnitude distributions. The synthetic waveform data set is made available online to facilitate future testing. Completeness strongly varies from the injection to the postinjection phase. The machine learning-enabled processing tool performs well at identifying S phases, quantifying event locations and magnitudes, dealing with site noise, and reliably retrieving all input events at magnitudes > −0.25. The real-time setup rapidly provides accurate estimates of the b value, robust characterization of large events, and approximate source information. Our study also provides insights into the feasibility of a digital “twin” component of wave propagation in an EGS. For time-invariant setups like the one presented, the generation of synthetic data is computationally feasible, whereas frequent or operational changes of the structure model would likely require surrogate modeling. Near-real-time source characterization of observed induced events, including focal mechanisms, could greatly enhance the value of geothermal reservoir digital twin components.
Stress Heterogeneities Governed by Fault Structure and Stress Transfer: The 2016–2017 Central Italy Seismic Sequence
Martínez-Garzón, Patricia; Meier, Men‐Andrin; Collettini, Cristiano; et al. (2025)
Journal of Geophysical Research: Solid Earth
We analyze the evolution of stress parameters from the 2016–2017 central Italy seismic sequence taking advantage of ∼13,747 robust focal mechanisms from a deep learning catalog. The density of the catalog allows us to invert focal mechanisms over distances of a few km and different time periods. We inferred a number of stress-related parameters, including the fault plane variability, the orientation of principal stress axes and maximum horizontal stress, the relative magnitudes of principal stresses and the variability of the principal stress orientations with respect to the median. From the uniform regional stress field consistent with the extension of the Apennine Belt, we observe local stress heterogeneities that are driven by the structural features and the coseismic stress history. A variation of the principal stress magnitudes and regimes from pure normal faulting toward transtension with depth is observed. Stress differences at the 1–10 km wavelength are observed between each side of two of the main regional fault structures. The reported stress results suggest a partial mechanical coupling and a strong interaction between the shallow normal faults and the detachment horizon at depth. Furthermore, distinct trends are observed in the stress parameters after the largest mainshocks, and before the M_W 6.5 Norcia mainshock, potentially indicating the high shear stress still available in well oriented faults after the M_W 6.0 Amatrice earthquake. Our analysis holds implications toward (a) constraining stress magnitudes, (b) illuminating the interaction between the shallow normal faults and detachment horizons, and (c) tracking stress evolution during seismic sequence.
Testing advanced seismic monitoring workflows at the Utah FORGE EGS site
Lanza, Federica; Shi, Peidong; Ermert, Laura; et al. (2025)
Proceedings of the European Geothermal Congress 2025
Many countries worldwide are investigating the potential of deep geothermal energy (DGE) as they plan to address future energy needs with more renewable solutions. However, the widespread development of DGE systems is challenged in many nations because of closely interrelated concerns about induced earthquakes, which can impact societal acceptance and economic viability due to inadequate production rates and high costs. Seismic monitoring plays a vital role in risk assessment and mitigation of DGE projects. However, increasing volumes of seismological data in recent years, requires new monitoring techniques to handle large and diverse datasets in near real-time and largely automated fashion.
Towards real-time forecasting of induced seismicity: demonstration at the Utah FORGE site
Ritz, Vanille; Lanza, Federica; Rinaldi, Antonio Pio; et al. (2025)
Proceedings of the European Geothermal Congress 2025
Enhanced Geothermal Systems (EGS) offer significant potential for low-carbon energy, but face challenges due to injection-induced seismicity. The Utah FORGE site, a flagship demonstration project funded by the U.S. Department of Energy, is designed to test and showcase the safe and effective deployment of EGS technologies. With its dense network of geophysical instrumentation and comprehensive monitoring capabilities, FORGE provides a unique environment for advancing real-time risk mitigation strategies. During hydraulic stimulations in 2022 and 2024, we implemented an Adaptive Traffic Light System (ATLS) that integrated three seismicity forecasting models in order to provide decision support to the operations. Our results show that combining diverse model classes enhances seismic hazard forecasting, supporting the safe and scalable adoption of EGS in the U.S., Switzerland, and beyond.
A Prospective Evaluation of Borehole Distributed Acoustic Sensing Technologies for Detecting and Locating Out‐of‐Network Microseismicity
Lanza, Federica; Tuinstra, Katinka B.; Rinaldi, Antonio P.; et al. (2024)
Geophysical Monograph Series ~ Distributed Acoustic Sensing in Borehole Geophysics
Downhole distributed acoustic sensing (DAS) has proven effective in monitoring microseismicity produced during hydraulic stimulations. It is also a promising tool for monitoring microseismicity occurring beyond the stimulation region. Downhole DAS arrays exhibit reduced sensitivity to surface noise compared to surface deployments, and they can remain in place undisturbed after operations cease, unlike downhole geophones. Identifying and locating microseismicity occurring at remote distances is crucial for reservoir monitoring, allowing us to assess stress conditions and the presence of pre-existing faults. Here, we evaluate the potential of downhole DAS in detecting and locating out-of-network microseismicity through two case studies: a field-scale investigation at the Frontier Observatory for Research in Geothermal Energy (FORGE) site in Utah and a mesoscale experiment in the Bedretto Underground Laboratory for Geosciences and Geoenergies (BULGG) in Switzerland. We demonstrate that DAS data have sufficient quality to detect and locate low-magnitude earthquakes (M < 0) occurring several hundred meters away. However, azimuthal ambiguity is unavoidable when only a single vertical borehole is available. We show that DAS signal characteristics and location capabilities perform equally at different scales, ranging from full-reservoir scale to mesoscale. Our evaluation supports the employment of longer-term experiments to explore downhole DAS suitability for passive microseismic monitoring.

Publications 1 - 10 of 12

Federica Lanza

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results