New Self-Supervised Foundation Model Improves Analysis of Global Satellite Displacement Data
Scientists have introduced GNSS-FM, a self-supervised foundation model pretrained on data from over 17,000 globally distributed GNSS stations to analyze daily displacement time series. The model adapts a masked latent prediction approach from audio processing (wav2vec 2.0) to geodetic data, using dual-stream inputs of displacement and velocity-like increments. It outperforms task-specific baselines on both 90-day displacement forecasting and seismic step localization, suggesting self-supervised pretraining can help overcome the scarcity of labeled GNSS data.
GNSS-FM is a newly proposed foundation model designed to process daily displacement time series from Global Navigation Satellite Systems, which are widely used to monitor tectonic crustal deformations and study earthquake cycles. A persistent challenge in applying machine learning to GNSS data has been the scarcity of labeled examples, despite large volumes of freely available unlabeled observations. To address this, the authors adapted a self-supervised pretraining strategy — masked latent prediction with vector-quantized targets — originally developed for speech processing in wav2vec 2.0, modifying it for geodetic time series. The model was pretrained on data from more than 17,000 globally distributed stations, and analysis of its learned internal representations indicates it captures key signal types including seismic offsets, tectonic drift, and seasonal patterns. When fine-tuned on downstream tasks, GNSS-FM outperformed strong supervised baselines in both 90-day displacement forecasting and seismic step localization. The work was submitted to arXiv in June 2026 and has not yet undergone formal peer review.
What's missing
As a preprint, the paper has not yet been peer-reviewed. The authors do not report ablation results comparing performance across different geographic or tectonic settings, leaving open questions about generalization to underrepresented regions. The scale of labeled fine-tuning data used and the computational cost of pretraining are not detailed in the abstract, limiting reproducibility assessment. Long-term performance stability and robustness to data gaps or sensor noise are not addressed.
What different sources said
- arXiv cs.LGCenter
GNSS-FM: A Self-Supervised Foundation Model for Daily GNSS Displacement Time Series
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.