GNSS-FM: A Self-Supervised Foundation Model for Daily GNSS Displacement Time Series

2026-06-05

AI Summary

This work addresses the scarcity of labeled data and overreliance on fully supervised learning in GNSS displacement time series analysis by introducing GNSS-FM, the first self-supervised foundation model for this domain. GNSS-FM employs a dual-stream input architecture—processing both displacement and pseudo-velocity increments—and leverages masked latent prediction with vector-quantized encoding. Pretrained on over 17,000 unlabeled global GNSS stations, the model effectively captures key geophysical signals, including coseismic offsets, tectonic drift, and seasonal variations. Evaluated on 90-day displacement forecasting and coseismic step detection tasks, GNSS-FM substantially outperforms strong baselines, demonstrating the efficacy and generalization capability of self-supervised pretraining for geodetic representation learning.

Abstract

Displacement time series from Global Navigation Satellite Systems (GNSS) are essential for a wide range of applications, including monitoring tectonic crustal deformations and investigating the different stages of the earthquake cycle. Machine learning methods have proven promising for GNSS applications; however, most remain fully supervised. This creates a bottleneck as labeled data are scarce, even though large amounts of unlabeled GNSS data are freely available. We present GNSS-FM, a self-supervised foundation model for daily GNSS time series. The model uses a dual-stream input combining displacement and velocity-like increments, and is pretrained using a masked latent prediction objective with vector-quantized targets adapted from wav2vec 2.0, with several modifications for geodetic data. Pretrained on data from over 17,000 globally distributed GNSS stations, an analysis of the learned codebook suggests that the representations capture the main signal types in GNSS displacement data, including seismic offsets, tectonic drift, and seasonal patterns. The foundation model is later fine-tuned on two downstream tasks, namely 90-day displacement forecasting and seismic step localization, where it outperforms strong task-specific baselines in both cases. These results show that self-supervised pretraining is a promising approach for GNSS time series analysis.

Problem

• GNSS
• displacement time series
• self-supervised learning
• labeled data scarcity
• foundation model

Innovation

• self-supervised learning
• foundation model
• GNSS time series
• masked latent prediction
• vector quantization