Machine Learning Model Improves Localized Earthquake Hazard Prediction Using Seismic Features and Spatial Data
Researchers have developed a spatiotemporal seismic hazard assessment framework that couples seismic statistical features with a VQ-VAE deep learning model trained on 2D seismic maps, improving localized earthquake prediction in Japan. The work extends a prior study that showed 60 seismic statistical features outperformed 428 generic time series features, now adding a spatially-aware neural component to predict magnitude 5.0+ events within a 24 km radius over a 15-day window. The approach is notable for producing a novel spatial feature that, according to SHAP analysis, largely replaces the traditionally used b-value in feature importance rankings.
A new preprint posted to arXiv presents a machine learning framework for localized seismic hazard assessment that combines seismic statistical features (SSFs) with a Vector Quantized Variational Autoencoder (VQ-VAE) trained on two-dimensional seismic maps. Building on prior work using XGBoost with earthquake catalogue data from Japan and Chile, the authors shift from whole-region to localized predictions, restricting both feature computation and prediction zones to a 24 km radius around candidate events. The VQ-VAE model is used to reconstruct seismic maps, with its reconstruction error interpreted as a proxy for localized crustal stress buildup — a novel spatial feature derived from 2D data rather than traditional 1D catalogue inputs. SHAP analysis ranks this VQ-VAE-derived feature at the top of feature importance, and the authors report it nearly entirely supplants the conventional b-value in the model's decision-making. Test AUC values for localized prediction remain comparable to those achieved in the previous whole-region Japan study, suggesting the localization does not degrade performance. The study relies exclusively on Japan data for the deep learning component, as the large dataset size is necessary for training the autoencoder. The paper was submitted in June 2026 and has undergone a minor title revision with no content changes.
What's missing
The study does not report external validation on held-out geographic regions beyond Japan, leaving generalizability uncertain. The physical interpretation of VQ-VAE reconstruction error as a crustal stress proxy is asserted but not independently validated against geodetic or geophysical stress measurements. Operational false-positive and false-negative rates at decision thresholds relevant to real-world hazard warning systems are not discussed, nor is comparison against established operational forecasting benchmarks such as CSEP.
What different sources said
- arXiv cs.LGCenter
Spatiotemporal Seismic Hazard Assessment Using VQ-VAE and Seismic Statistical Features
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.