Montzka et al. (2025) AI in soil moisture remote sensing

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: International Journal of Applied Earth Observation and Geoinformation
• Year: 2025
• Authors: Carsten Montzka, Luca Brocca, Hao Chen, Narendra N. Das, Antara Dasgupta, Mehdi Rahmati, Thomas Jagdhuber
• DOI: 10.1016/j.jag.2025.105011

Research Groups

• Not specified in the provided text (Review article).

Short Summary

This paper provides the first structured overview of artificial intelligence (AI) applications for soil moisture retrieval from remote sensing data. It highlights how AI overcomes the limitations of traditional physical models by learning complex non-linear relationships and improving data continuity and resolution.

Objective

• To critically review emerging AI-based soil moisture retrieval methods, evaluate their advantages and disadvantages, and identify future research directions to optimize the use of novel sensors and data.

Study Configuration

• Spatial Scale: Global and regional (remote sensing applications).
• Temporal Scale: Not specified (covers time series reconstruction and forecasting).

Methodology and Data

• Models used: Various Artificial Intelligence (AI) and Machine Learning (ML) architectures, including those for time series reconstruction, spatial scaling, and predictive forecasting.
• Data sources: Satellite-based remote sensing observations, ground reference data (in situ measurements), and auxiliary hydrological/environmental inputs.

Main Results

• AI models effectively capture highly non-linear relationships between satellite signals and ground-based soil moisture references.
• AI facilitates time series reconstruction by accurately filling gaps in satellite datasets.
• The technology enables the estimation of subsurface soil moisture conditions using only surface signals and auxiliary data.
• AI supports spatial scaling, allowing for the translation of soil moisture estimates across different spatial resolutions using multi-resolution data.
• Data-driven methods provide robust temporal dynamics predictions for soil moisture forecasting and broader water cycle assessments.

Contributions

• Provides the first comprehensive and structured overview of AI's role in remote sensing-based soil moisture retrieval.
• Identifies specific scientific challenges and future research directions to guide the community in the era of expanding AI applications.
• Demonstrates the superiority of data-driven approaches over traditional methods in handling model uncertainties and sensor-related constraints.

Funding

• Not specified in the provided text.

Citation

@article{Montzka2025AI,
  author = {Montzka, Carsten and Brocca, Luca and Chen, Hao and Das, Narendra N. and Dasgupta, Antara and Rahmati, Mehdi and Jagdhuber, Thomas},
  title = {AI in soil moisture remote sensing},
  journal = {International Journal of Applied Earth Observation and Geoinformation},
  year = {2025},
  doi = {10.1016/j.jag.2025.105011},
  url = {https://doi.org/10.1016/j.jag.2025.105011}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)