We investigate the application of ultra-high frequency (UHF) seismic for soil analysis at the decimetre scale. We conducted experiments in agricultural fields and controlled environments to assess the efficacy of various seismic sources, receiver types, and data-processing approaches. Our experiments demonstrate that a coherent UHF (>500Hz) seismic wavefield can be recorded by 16 ground-motion sensors over a distance of 3m.

We identify UHF P-waves, S-waves, and surface waves with clear move-out. Hammer strikes generate high-amplitude and impulsive signals in the 800-1500Hz range. Among tested receivers, the LOM geophone (priced under £100) has adequate sensitivity for field deployments, and is a low-cost alternative to industry-standard accelerometers. 

This opens the door to applying conventional imaging techniques such as first-arrival tomographic imaging, surface-wave dispersion analysis, full-waveform inversion, or machine-learning based inversions for estimating the properties of top soil relevant to farmers.

Understanding soil is fundamental to agriculture, carbon cycling, and environmental sustainability, yet progress is limited by fragmented and heterogeneous datasets that constrain modeling to small-scale predictive settings rather than high-dimensional representation learning. We introduce LUCAS-MEGA, a large-scale multimodal dataset constructed through systematic data fusion of European soil--environment observations, with the LUCAS survey as its backbone. This fusion dataset comprises over 70,000 samples and more than 1,000 features spanning physical, chemical, environmental, and visual properties, aggregated from 68 source datasets. To enable integration at scale, we develop SoilFuser, a multi-agent, human-in-the-loop data fusion pipeline that standardizes heterogeneous data formats and measurement protocols, resolves inconsistencies and invalid entries (e.g., unit and codebook mismatches and erroneous values), incorporates natural language annotations, and harmonizes multimodal attributes and metadata into a unified, machine learning-ready feature space. The resulting dataset captures key characteristics of real-world soil observations, including multi-modality, uneven feature coverage, and heterogeneous uncertainty. To demonstrate the usability of LUCAS-MEGA for data-driven modeling, we pretrain a multimodal tabular transformer (SoilFormer) using a self-supervised objective based on feature masking, achieving stable training, strong predictive performance, and uncertainty-aware representations. We further show that the learned representations recover relationships consistent with established soil processes. LUCAS-MEGA is released with open access and is accompanied by composable, agent-friendly APIs that support structured querying and data-driven workflows.