A Three-Dimensional FDTD(2,4) Subgridding Algorithm for the Airborne Ground-Penetrating Radar Detection of Landslide Models
Remote Sensing · 20.03.2025 · 17 (6) · 1107
DOI: 10.3390/rs17061107 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 20.03.2025. Verfasst von Lifeng Mao und 5 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
The finite-difference time-domain (FDTD) method is a robust numerical approach for the three-dimensional forward modeling of airborne ground-penetrating radar responses of complex geological structures, particularly landslides. However, standard FDTD implementations encounter significant memory demands as aircraft altitude increases and when modeling high-permittivity subsurface media (e.g., water-saturated soils), often exceeding ordinary computational resources. Existing subgridding FDTD methods, tailored for simple localized target models, are also inadequate for simulating landslide models. To overcome these limitations, we thus propose a novel high-order FDTD-based subgridding algorithm that applies coarse grids to the air layer and fine grids to the subsurface medium, enabling the simulation of arbitrarily complex landslide models with significantly reduced memory consumption. This study achieves the first implementation of the high-order FDTD(2,4) method in both coarse- and fine-grid regions, which enables larger grid sizes in both regions. As a result, the proposed approach not only preserves high-order spatial accuracy but also achieves significant memory savings. To mitigate the challenges posed by higher-order difference stencils, we introduce a specialized grid configuration with an overlap zone between coarse and fine grids, supplemented by surrounding virtual nodes. The algorithm accommodates various grid refinement factors, ensuring adaptability to dielectric models with diverse permittivity values and structural complexities. By optimizing the grid refinement factor based on the subsurface medium’s maximum permittivity, simulations can be performed with minimal memory usage. Field updates within the overlapping region are followed by weighted corrections to ensure numerical stability, whereas simulations without these novel measures exhibit oscillatory artifacts. Wavefield snapshots reveal seamless transitions across grid boundaries without spurious artifacts. Numerical experiments on deposition-type landslide models and water-bearing media confirm the validity and stability of the proposed method. Notably, using the optimal grid refinement factor reduces memory consumption to less than 8% of the standard FDTD method for aquifer model simulations.
Förderung
- National Key R&D Program of China
- National Natural Science Foundation of China
Zitieren
APA
Mao, L., Wang, X., Chi, Y., Pang, S., Wang, X., & Huang, Q. (2025). A Three-Dimensional FDTD(2,4) Subgridding Algorithm for the Airborne Ground-Penetrating Radar Detection of Landslide Models. Remote Sensing, 17(6), 1107. https://doi.org/10.3390/rs17061107
Vancouver
Mao L, Wang X, Chi Y, Pang S, Wang X, Huang Q. A Three-Dimensional FDTD(2,4) Subgridding Algorithm for the Airborne Ground-Penetrating Radar Detection of Landslide Models. Remote Sensing. 2025;17(6):1107. doi:10.3390/rs17061107
BibTeX
@article{Mao2025,
author = {Mao, Lifeng and Wang, Xuben and Chi, Yuelong and Pang, Su and Wang, Xiangpeng and Huang, Qilin},
title = {A Three-Dimensional FDTD(2,4) Subgridding Algorithm for the Airborne Ground-Penetrating Radar Detection of Landslide Models},
journal = {Remote Sensing},
year = {2025},
volume = {17},
number = {6},
pages = {1107},
doi = {10.3390/rs17061107}
}
EUVIMED