Spatiotemporal Modeling of Soil Moisture in Humid Areas by Integrating Transformer Architecture and Remote Sensing Data
Main Article Content
Abstract
Soil moisture variation in humid regions presents high-frequency nonlinearity and is strongly influenced by surface conditions, making it difficult for traditional time series models to effectively capture long-range dependencies. In this study, a multimodal Transformer network architecture is proposed, integrating Sentinel-1 VV/VH radar data, MODIS NDVI, and meteorological variables to predict daily 0–10 cm soil moisture content at a 1 km² grid scale in Jiangning District, Nanjing. Data from 2021 were used for model training, and 2022 data were used for testing. Verified by in-situ measurements, the monthly average of 0–10 cm soil moisture in Jiangning during 2021 ranged from 0.20 to 0.32 m³/m³, with a standard deviation of 0.035 m³/m³, reflecting its high-frequency nonlinear characteristics. The model achieved an average RMSE of 0.022 m³/m³, which was lower than that of LSTM (RMSE = 0.029) and traditional SVR (RMSE = 0.034). The model's interpretability module (attention map) showed that vegetation cover and rainfall in the previous six days contributed 41.2% and 27.8%, respectively. This study provides an AI-based approach for modeling soil–vegetation–hydrology interactions driven by remote sensing data.
Article Details
Issue
Section
How to Cite
References
1. X. Wang, R. Liu, M. Köhli, J. Marach, and Z. Wang, "Monitoring soil water content and measurement depth of cosmic-ray neutron sensing in the Tibetan Plateau," Journal of Hydrometeorology, vol. 26, no. 2, pp. 155–167, 2025.
2. Y. Xiao, L. Tan, and J. Liu, "Application of Machine Learning Model in Fraud Identification: A Comparative Study of CatBoost, XGBoost and LightGBM," 2025.
3. G. Jiang, J. Yang, S. Zhao, H. Chen, Y. Zhong, and C. Gong, "Investment Advisory Robotics 2.0: Leveraging Deep Neural Networks for Personalized Financial Guidance," 2025.
4. J. Li, S. Wu, and N. Wang, "A CLIP-Based Uncertainty Modal Modeling (UMM) Framework for Pedestrian Re-Identification in Autonomous Driving," 2025.
5. J. Liu et al., "Analysis of collective response reveals that COVID-19-related activities start from the end of 2019 in mainland China," medRxiv, 2020-10, 2020.
6. J. Tian, J. Lu, M. Wang, H. Li, and H. Xu, "Predicting Property Tax Classifications: An Empirical Study Using Multiple Machine Learning Algorithms on US State-Level Data," 2025.
7. Y. Wang, X. Han, and X. Zhang, "AI-Driven Market Segmentation and Multi-Behavioral Sequential Recommendation for Personalized E-Commerce Marketing," 2025.
8. T. Yuan, X. Zhang, and X. Chen, "Machine Learning based Enterprise Financial Audit Framework and High Risk Identification," arXiv preprint arXiv:2507.06266, 2025.
9. X. Wang, Z. L. Wang, W. Yang, and R. Liu, "Hydraulic properties within the complete moisture range of hydric soil on the Tibetan Plateau," Water Resources Research, vol. 60, no. 7, e2023WR036018, 2024.
10. J. Yang, "Predicting water quality through daily concentration of dissolved oxygen using improved artificial intelligence," Scientific Reports, vol. 13, no. 1, p. 20370, 2023.
11. H. Gui, Y. Fu, B. Wang, and Y. Lu, "Optimized Design of Medical Welded Structures for Life Enhancement," 2025.
12. F. Chen, H. Liang, L. Yue, P. Xu, and S. Li, "Low-Power Acceleration Architecture Design of Domestic Smart Chips for AI Loads," 2025.
13. F. Zhang, "Distributed Cloud Computing Infrastructure Management," International Journal of Internet and Distributed Systems, vol. 7, no. 3, pp. 35–60, 2025.
14. Y. Lin, Y. Yao, J. Zhu, and C. He, "Application of Generative AI in Predictive Analysis of Urban Energy Distribution and Traffic Congestion in Smart Cities," in Proc. 2025 IEEE International Conference on Electronics, Energy Systems and Power Engineering (EESPE), pp. 765–768, Mar. 2025.
15. H. Gui, B. Wang, Y. Lu, and Y. Fu, "Computational Modeling-Based Estimation of Residual Stress and Fatigue Life of Medical Welded Structures," 2025.
16. M. Yang, Y. Wang, J. Shi, and L. Tong, "Reinforcement Learning Based Multi-Stage Ad Sorting and Personalized Recommen-dation System Design," 2025.
17. K. Xu, X. Mo, X. Xu, and H. Wu, "Improving Productivity and Sustainability of Aquaculture and Hydroponic Systems Using Oxygen and Ozone Fine Bubble Technologies," Innovations in Applied Engineering and Technology, pp. 1–8, 2022.
18. Y. Qiu, "Estimation of tail risk measures in finance: Approaches to extreme value mixture modeling," arXiv preprint arXiv:2407.05933, 2024.
19. S. Zhan, Y. Lin, J. Zhu, and Y. Yao, "Deep Learning Based Optimization of Large Language Models for Code Generation," 2025.
20. H. Chen, J. Li, X. Ma, and Y. Mao, "Real-Time Response Optimization in Speech Interaction: A Mixed-Signal Processing Solu-tion Incorporating C++ and DSPs," SSRN 5343716, 2025.
21. R. Liang, Z. Ye, Y. Liang, and S. Li, "Deep Learning-Based Player Behavior Modeling and Game Interaction System Optimiza-tion Research," 2025.
22. H. Peng, N. Dong, Y. Liao, Y. Tang, and X. Hu, "Real-Time Turbidity Monitoring Using Machine Learning and Environmental Parameter Integration for Scalable Water Quality Management," Journal of Theory and Practice in Engineering and Technology, vol. 1, no. 4, pp. 29–36, 2024.
23. J. Zheng and M. Makar, "Causally motivated multi-shortcut identification and removal," Advances in Neural Information Pro-cessing Systems, vol. 35, pp. 12800–12812, 2022.
24. K. Xu, X. Xu, H. Wu, and R. Sun, "Venturi Aeration Systems Design and Performance Evaluation in High Density Aquaculture," 2024.
25. K. Xu, X. Xu, H. Wu, R. Sun, and Y. Hong, "Ozonation and Filtration System for Sustainable Treatment of Aquaculture Wastewater in Taizhou City," Innovations in Applied Engineering and Technology, pp. 1–7, 2023.