Science AI Journal
Computer Science · Research topic

Open research questions in Seismology and Earthquake Studies

54 unresolved questions extracted from the limitations and future-work sections of 127 Seismology and Earthquake Studies papers in our library. Each links back to the study that raised it.

What the literature leaves open

  • Complex near-surface conditions. Strong noise contamination. Weak signal energy. Limited accuracy and poor continuity of conventional automatic picking methods.

    Research on an automatic seismic first-arrival picking application based on deep learning semantic segmentation · 2026 · DOI

  • The study relies on synthetic seismic datasets, which may not fully represent real-world scenarios. The framework is evaluated on a limited number of datasets. The study does not provide a comprehensive comparison with other existing methods.

    Research on an automatic seismic first-arrival picking application based on deep learning semantic segmentation · 2026 · DOI

  • The proposed method's primary boundary lies in ultralow SNR conditions (below approximately -26 dB). Robustness against impulsive noise remains unverified due to its sparsity similarity to effective signals.

    Research on microseismic data noise suppression method based on adaptive spectral segmentation · 2026 · DOI

  • Existing methods generally suffer from poor adaptability, low processing accuracy, and insufficient computational efficiency. The development of adaptive denoising techniques has become an important research direction in this field.

    Research on microseismic data noise suppression method based on adaptive spectral segmentation · 2026 · DOI

  • The earthquake posed significant challenges to emergency responders and infrastructure managers. The earthquake highlighted the need for effective communication and coordination between government agencies and emergency responders. The earthquake demonstrated the importance of rapid functional recovery strategies.

    The 2024 Hualien, Taiwan earthquake: NZSEE learning from earthquakes reconnaissance report · 2026 · DOI

  • Fragmented and rarely cross-validated information. Scarcity of harmonized event data. Limited implementation of multi-hazard perspectives at local scales.

    Natural hazard events dataset for the Garrotxa Region (Catalonia, Spain): a foundational step toward multi-hazard risk assessment · 2026 · DOI

  • Further research on the application of foundation models in geophysics. Exploration of other data augmentation techniques and clustering strategies. Investigation of the potential applications of the proposed SPFM paradigm in other fields.

    Light-Weighted Foundation Model for Seismic Data Processing Based on Representative and Non-redundant Pre-training Dataset · 2026 · DOI

  • The lack of a technological framework for foundation models in geophysics. The scarcity of publicly available datasets for seismic data processing. The high computational cost associated with large model parameters.

    Light-Weighted Foundation Model for Seismic Data Processing Based on Representative and Non-redundant Pre-training Dataset · 2026 · DOI

  • The study uses a single building as a case study. The model requires a significant amount of data to train. The study does not address the issue of structural damage detection.

    Machine learning–driven structural health monitoring of a high-rise building on thick sediments via seismic ambient noise · 2026 · DOI

  • Further studies can investigate the use of seismic ambient noise for structural health monitoring of other types of buildings. The development of more advanced machine learning models can improve the accuracy of predictions. The study's findings can be used to develop more effective strategies for mitigating the risks of structural damage due to seismic site amplification effect.

    Machine learning–driven structural health monitoring of a high-rise building on thick sediments via seismic ambient noise · 2026 · DOI

  • Low signal-to-noise ratio conditions. Similarity between blast and earthquake seismic signals. Limited station coverage in some regions.

    Effectively Distinguishing Blast and Earthquake Sources in Eastern Canada · 2026 · DOI

  • Although machine learning-based data processing techniques have significantly improved data repeatability, research on quantifying the uncertainty in the predictions generated by these models remains insufficient.

    Uncertainty Quantification of Machine Learning-Based Repeatability Enhancement for Time-lapse Seismic Data · 2026 · DOI

  • While the method identifies time-frequency features in the 30-40 Hz range for foreshocks, the physical mechanisms explaining why these specific frequencies are indicative of foreshock activity remain to be fully established.

    Analysis of Foreshocks and Aftershocks in a microseismic sequence in Switzerland using Explainable AI · 2026 · DOI

  • Ongoing developments include spectral moment magnitudes and DAS strain-rate-based magnitude estimates to improve characterization of small-magnitude events.

    Multi-Scale On-Fault Seismic Monitoring at the Bedretto Underground Laboratory: An Operational Framework · 2026 · DOI

  • Latency characterization isolates preprocessing overhead and TensorRT FP16 forward pass on a single GPU class, but field systems vary from embedded devices to shared servers; reproducing measurements under representative I/O and telemetry loads remains essential.

    Real-time seismic signal classification using deep vision architectures for earthquake early warning systems · 2026 · DOI

  • Missed detections concentrate in low-SNR regimes; three avenues are suggested: (i) SNR-aware decision rules with higher thresholds or corroboration, (ii) targeted augmentation and curriculum sampling emphasizing emergent low-amplitude onsets, and (iii) multi-station fusion within short spatial apertures to boost effective SNR.

    Real-time seismic signal classification using deep vision architectures for earthquake early warning systems · 2026 · DOI

  • This study proposed a data-driven machine learning framework for rockburst prediction and seismic resilience assessment through probabilistic fragility analysis. The workflow integrates preprocessing and feature engineering, ensemble learning, performance evaluation, explainable AI, and construction of a threedimensional (3D) fragility surface to translate predictive outputs vulnerability representation. By combining high-performing learners within a voting-based ensemble and linking predictions to physically meaningful seismic the framework supports both accurate hazard prediction and practical decision support for underground operations.

    Data-Driven Explainable Machine learning for Rockburst Hazard and Seismic Resilience with 3D Fragility Surfaces · 2026 · DOI

  • (b) LIME explanation for Seismic prediction Figure 14: Model interpretability using LIME for two instances representing different prediction classes. (Developed by Authors) The LIME explanations show how individual features contribute to a specific prediction. For cases classified as hazardous, energy- and acceleration-related features typically contribute positively toward the rockburst class, while other predictors may reduce hazard probability depending on their observed values. This instance-level interpretability is useful for operational deployment because to understand why an alert is triggered for a particular event and whether the explanation is consistent with engineering judgment. it enables practitioners Figure 15: SHAP explanation for feature contribution analysis. (Developed by Authors) SHAP provides a global interpretation by quantifying feature contributions across the dataset. Consistent with the feature-importance results, SHAP indicates that seismic energy and magnitude-related measures play dominant roles in shifting predictions toward higher risk. Importantly, combining SHAP (global) and LIME identifies overall trust: SHAP (local) strengthens drivers, while LIME explains event-specific triggers. This dual interpretability supports practical engineering requirements and improves model credibility for decision support.

    Data-Driven Explainable Machine learning for Rockburst Hazard and Seismic Resilience with 3D Fragility Surfaces · 2026 · DOI

  • Further analysis of the seismicity in the Gargano area using the new catalog. Application of the CASP software to other seismic networks to improve earthquake detection capabilities. Investigation of the seismotectonic and seismic hazard characteristics in the Gargano area.

    The new seismic catalog of the Gargano area (Southern Italy) after a decade of seismic monitoring by OTRIONS network · 2026 · DOI

  • The first available seismic catalog suffered from technological problems. The need for a new seismic catalog that includes the most recent seismicity and covers the temporal gaps existing in the previous catalog.

    The new seismic catalog of the Gargano area (Southern Italy) after a decade of seismic monitoring by OTRIONS network · 2026 · DOI

  • Applying the framework to other intraplate regions. Comparing the framework with other seismic source discrimination methods. Improving the framework to handle low signal-to-noise ratio conditions.

    Effectively Distinguishing Blast and Earthquake Sources in Eastern Canada · 2026 · DOI

  • Further testing of ORION on different datasets and hardware configurations. Exploration of the application of ORION to other types of seismic data.

    Near real-time channel selection for Distributed Acoustic Sensing technology · 2026 · DOI

  • The lack of efficient processing strategies capable of handling high data complexity both in terms of size and signal quality. The need for a near real-time selector of high-quality DAS channels.

    Near real-time channel selection for Distributed Acoustic Sensing technology · 2026 · DOI

  • The lack of integration of heterogeneous spatial data for earthquake disaster management is a significant gap. The need for advanced data mining and Business Intelligence techniques for spatial data integration is also a gap.

    Integration of Spatial Data from Heterogeneous Sources and Their Handling in a Dashboard for Earthquake Disaster Management · 2026 · DOI

  • This study explored the integration of spatial and non-spatial data to design an intelligent, location-aware dashboard that supports proactive earthquake disaster management. In addressing the core research questions including how to implement such a dashboard, what features are essential and how effective data integration can be in seismic crisis response, a practical prototype was developed using Power BI enhanced by Python-based ETL processes and ML. The research confirms the first hypothesis, asserting that while urban and seismic datasets are often fragmented, they are nonetheless of sufficient quality to support reliable seismic vulnerability modeling. The use of Random Forest algorithm enabled accurate classification of urban blocks into seismic vulnerability zones (very low vulnerability, low vulnerability, medium vulnerability, high vulnerability and very high vulnerability). The dashboard developed in this study offers several unique advantages: Localization and High Spatial Accuracy: Unlike generalized international dashboards, this system is specifically designed for Tehran Municipality District 2, providing detailed insights at the urban block level. Scenario-Based Interactivity: While most existing dashboards are static and data-display oriented, this dashboard supports interactive analysis based on hypothetical earthquake scenarios defined by the user. Integration with Local Government Infrastructure: Built with Power BI, the dashboard can be seamlessly integrated into organizational Information Technology (IT) systems and workflows commonly used by local agencies. The proposed dashboard demonstrates significant potential as an innovative and practical tool for: Urban planners, to prioritize building retrofitting projects, disaster managers, to simulate earthquake impacts and assess various risk scenarios and non-technical decision-makers, who benefit from simplified visuals such as maps and charts that enhance understanding and situational awareness. This research contributes to the field through: Embedding live spatial modeling and ETL pipelines within Power BI: a rare combination in disaster analytics. Allowing interactive scenario simulations: users can input earthquake magnitude and depth and instantly visualize shifting vulnerability patterns. Demonstrating the potential of hybrid low-code platforms (Power BI + Python) to serve both technical and nontechnical stakeholders: reducing barriers to adoption. Providing a localized a model: that is focused on blocks and neighborhoods. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W8-2025 The 8th ISPRS Geospatial Conference 2025, 15–17 December 2025, Tehran, IranThis contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper. https://doi.org/10.5194/isprs-annals-X-4-W8-2025-1-2026 | © Author(s) 2026. CC BY 4.0 License.

    Integration of Spatial Data from Heterogeneous Sources and Their Handling in a Dashboard for Earthquake Disaster Management · 2026 · DOI

Most-cited papers in Seismology and Earthquake Studies

Most recent work

Find a gap in your own Seismology and Earthquake Studies sub-topic

This page shows what the Seismology and Earthquake Studies literature already flags as unresolved. To narrow it to your specific question, run the guided finder — it searches the gap library on demand and checks candidates against 250M+ OpenAlex works.

Open the Research Gap Finder →

Related topics in Computer Science

Online Learning and Analytics
851 open questions
Digital literacy in education
518 open questions
AI in Service Interactions
517 open questions
Music Technology and Sound Studies
423 open questions
Teaching and Learning Programming
384 open questions
Quantum Computing Algorithms and Architecture
368 open questions
Virtual Reality Applications and Impacts
317 open questions
Intelligent Tutoring Systems and Adaptive Learning
298 open questions
54 open questions have been extracted from the limitations and future-work passages of 127 Seismology and Earthquake Studies papers in our library. Each one below links back to the study that raised it, so you can read the original claim in context.

Tools for your next paper

Gaps & Literature SearchFind open questions in any field — 120,000+ open gaps mined from our own library.Pre-CheckIs your paper ready? A Tier 1–5 acceptance probability in seconds.Journal FinderA ranked shortlist of target journals from a 17,500-venue index.Duplicate Publication CheckerCheck prior-publication & salami-slicing risk across 8 sources.Citation GeneratorFreeA DOI or title → APA, MLA, Vancouver, Chicago, BibTeX, RIS. No signup.Graphical Abstract MakerFreeTurn your findings into a colorblind-safe graphical abstract.

Compare the category — Honest roundups of the AI research tools, ours listed alongside the alternatives.

Command palette

Jump anywhere, run any action.