A multimodal Transformer for InSAR-based ground deformation forecasting with cross-site generalization across Europe

TL;DR

This paper introduces a multimodal Transformer model for predicting ground deformation from InSAR data, demonstrating superior accuracy and generalization across European sites by integrating recent displacement, static indicators, and harmonic encodings.

Contribution

The study presents a novel multimodal Transformer architecture that improves next-epoch ground displacement forecasting from InSAR time series with cross-site generalization capabilities.

Findings

Outperforms CNN-LSTM and other models in displacement prediction accuracy.

Achieves RMSE of 0.90 mm and R^2 of 0.97 on test data.

Effective in modeling complex deformation patterns across different sites.

Abstract

Near-real-time regional-scale monitoring of ground deformation is increasingly required to support urban planning, critical infrastructure management, and natural hazard mitigation. While Interferometric Synthetic Aperture Radar (InSAR) and continental-scale services such as the European Ground Motion Service (EGMS) provide dense observations of past motion, predicting the next observation remains challenging due to the superposition of long-term trends, seasonal cycles, and occasional abrupt discontinuities (e.g., co-seismic steps), together with strong spatial heterogeneity. In this study we propose a multimodal patch-based Transformer for single-step, fixed-interval next-epoch nowcasting of displacement maps from EGMS time series (resampled to a 64x64 grid over 100 km x 100 km tiles). The model ingests recent displacement snapshots together with (i) static kinematic indicators (mean velocity, acceleration, seasonal amplitude) computed in a leakage-safe manner from the training window only, and (ii) harmonic day-of-year encodings. On the eastern Ireland tile (E32N34), the STGCN is strongest in the displacement-only setting, whereas the multimodal Transformer clearly outperforms CNN-LSTM, CNN-LSTM+Attn, and multimodal STGCN when all models receive the same multimodal inputs, achieving RMSE = 0.90 mm and $R^2$ = 0.97 on the test set with the best threshold accuracies.