VIs are critical for monitoring vegetation health and productivity using satellite imagery. However, soil background effects, especially variations in soil moisture and type, can introduce significant uncertainties, particularly in sparsely vegetated areas and coarse resolution imagery. These variations complicate accurate vegetation assessment by mixing vegetation and soil spectral signatures. Therefore, understanding how different indices resist these soil influences is essential for enhancing remote sensing data reliability. Based on these challenges, there is a need for deeper research into improving VIs' sensitivity to both soil moisture and type.
Published (DOI: 10.34133/remotesensing.0994) in Journal of Remote Sensing on January 8, 2026, this study investigates the sensitivity of various soil-resistant VIs to soil moisture and soil type. The research, conducted by the State Key Laboratory of Remote Sensing Science at Beijing Normal University and other institutions, provides an in-depth evaluation of 31 soil-resistant indices, focusing on their ability to minimize the uncertainty introduced by diverse soil backgrounds. The study is crucial for advancing remote sensing techniques, particularly in agricultural monitoring and ecological studies, where accurate vegetation data is essential.
The study categorized the indices into six groups based on their design principles, including soil-line adjusted, photosynthesis-oriented, shape separation, SWIR-adjusted, RedEdge-adjusted, and green triangular indices. Through a combination of 3D radiative transfer model simulations and ground-based experiments, the researchers assessed the performance of each VI under varying soil conditions. The study found that soil background effects significantly impact traditional vegetation indices like NDVI. Among the 31 soil-resistant indices, 22 (in simulation experiments) and 26 (in ground-based experiments) outperformed NDVI in mitigating soil effects.
Several soil-line adjusted, SWIR-adjusted and green triangular indices showed superior resistance to soil type and moisture variations. These indices are highly effective in regions with complex soil conditions, making them suitable for large-scale agricultural and ecological applications. RedEdge adjusted indices showed a higher sensitivity to soil type, while some SWIR-adjusted indices are sensitive to soil moisture. The findings emphasize that while some indices excel in specific conditions, a tailored approach is necessary for diverse soil types and moisture levels.
Dr. Cong Wang, one of the lead researchers, states, “Our findings are a significant step forward in improving remote sensing accuracy. By understanding the nuances of how soil moisture and type affect vegetation indices, we can make better-informed decisions in agricultural monitoring and environmental conservation.”
The study employed both simulation and real-world ground-based experiments to evaluate the sensitivity of different VIs to soil variations. The simulations were conducted using a 3D radiative transfer model (LESS), while field experiments involved collecting soil and vegetation data from various regions across China. The experiments considered multiple factors such as vegetation coverage, soil type, and moisture levels to assess the effectiveness of each index in mitigating soil effects.
The development of more robust soil-resistant indices opens the door for improved remote sensing applications in agriculture, especially in monitoring crop health and estimating biophysical parameters. These advancements could also enhance environmental monitoring, helping to track vegetation changes in areas impacted by climate variability. The future of remote sensing lies in integrating these advanced indices with global satellite systems to provide real-time, accurate data on vegetation dynamics across the globe.
###
References
DOI
10.34133/remotesensing.0994
Original Source URL
https://spj.science.org/doi/10.34133/remotesensing.0994
Funding Information
This study was supported by the grants from the National Key R&D Program of China (2022YFD2001103), the Innovation Program of Chinese Academy of Agricultural Sciences (CAAS-CSAL-202402), and the National Natural Science Foundation of China (42101370).
About Journal of Remote Sensing
The JJournal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.