From Image Morphology to Multiphysics Earth Posteriors
Abstract
Structural learning and physical-property inference can draw on different sources of knowledge. We learn morphology from 33,140 natural, satellite, and texture images, without geological training models, then infer Earth properties from physical observations. Identical frozen weights map 128 coordinates into each 256×256 compressional-velocity, shear-velocity, or density field. Fixed adapters assign physical meaning; observation-specific likelihoods determine new posteriors without retraining. Across 256 unseen Earth models, image coordinates reduce projection error fourfold relative to a dimension-matched cosine basis. Matched inversions favor image coordinates for the tested curved-velocity targets and a smooth radial basis for the tested faulted targets. A simplified 2.5-dimensional application with 915 measured observations along one 120.9-km southern-California profile constrains a velocity--density contrast. Grouped arrival-time and gravity errors fall by 43% and 44% relative to the pretrained image prior, although the matched smooth radial basis predicts better. Direct layered-Rayleigh updating reduces blind phase-velocity error by 28% at five sites, whereas P-and-gravity-only sharing worsens it, distinguishing reusable morphology from beneficial cross-property coupling. These results establish a practical separation: images supply structural alternatives, and heterogeneous physical observations determine their experiment-specific posterior probabilities.
Related
- Image Morphology of Misaligned Magnetically Arrested Disks
- Marginalized planet-to-star radius ratio posteriors for low-SNR transiting planets via dimensionality reduction
- Asteroid Factory: High-fidelity Multiphysics Modeling of Small Solar System Bodies via Volumetric Aggregate Synthesis
- Juice/SWI during the Lunar-Earth-Gravity-Assist (LEGA) – Part 3: Observations of the Earth as calibration target
- Magnetic connectivity from the Sun to the Earth with MHD models. II. Impact of input magnetic maps
BibTeX
@misc{yu2026from,
title = {{From Image Morphology to Multiphysics Earth Posteriors}},
author = {Ziye Yu and Xin Liu and Yuqi Cai},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.31229},
eprint = {2609.31229},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.31229},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - From Image Morphology to Multiphysics Earth Posteriors
AU - Ziye Yu
AU - Xin Liu
AU - Yuqi Cai
PY - 2026
DA - 2026/09/25
PB - arXiv
DO - 10.48550/arxiv.2609.31229
UR - https://arxiv.org/abs/2609.31229
AB - Structural learning and physical-property inference can draw on different sources of knowledge. We learn morphology from 33,140 natural, satellite, and texture images, without geological training models, then infer Earth properties from physical observations. Identical frozen weights map 128 coordinates into each 256×256 compressional-velocity, shear-velocity, or density field. Fixed adapters assign physical meaning; observation-specific likelihoods determine new posteriors without retraining. Across 256 unseen Earth models, image coordinates reduce projection error fourfold relative to a dimension-matched cosine basis. Matched inversions favor image coordinates for the tested curved-velocity targets and a smooth radial basis for the tested faulted targets. A simplified 2.5-dimensional application with 915 measured observations along one 120.9-km southern-California profile constrains a velocity--density contrast. Grouped arrival-time and gravity errors fall by 43% and 44% relative to the pretrained image prior, although the matched smooth radial basis predicts better. Direct layered-Rayleigh updating reduces blind phase-velocity error by 28% at five sites, whereas P-and-gravity-only sharing worsens it, distinguishing reusable morphology from beneficial cross-property coupling. These results establish a practical separation: images supply structural alternatives, and heterogeneous physical observations determine their experiment-specific posterior probabilities.
ER -