Advancing wetland vegetation monitoring: A novel spectral method to estimate key nutrient elements

By analyzing spectral data from multiple vegetation species in Southwest China, the research team developed a more accurate, efficient approach to estimate leaf nitrogen, phosphorus, and potassium content (LNC, LPC, and LKC).

Karst wetlands, with their distinctive calcium-rich soils and dynamic hydrology, face growing ecological challenges due to biodiversity loss, habitat degradation, and climate change. Accurate monitoring of vegetation health in these fragile environments is critical, as nitrogen, phosphorus, and potassium are essential nutrients influencing plant growth and stress resistance. Traditional methods of measuring these elements are labor-intensive, often requiring destructive sampling and time-consuming lab analysis. Remote sensing offers a non-destructive alternative, but previous methods struggled with limited spectral sensitivity and environmental constraints, particularly in karst wetlands.

A study (DOI: 10.1016/j.plaphe.2025.100120) published in Plant Phenomics on 13 October 2025 by Bolin Fu’s team, Guilin University of Technology, enhances remote sensing technologies for wetland ecosystem management, offering a path forward for monitoring vegetation health and maintaining ecosystem balance.

This study analyzed hyperspectral leaf data from seven vegetation species in a karst wetland ecosystem, using continuous wavelet decomposition to reveal unique spectral properties. Notably, water absorption bands at 1400 nm and 1900 nm shifted towards longer wavelengths, creating reflection peaks at 1655 nm and 2220 nm. Reflectance between 760–975 nm consistently exceeded 0.55, higher than typical green vegetation. The spectral decomposition across scales 1 to 7 improved clarity and differentiation among species. The researchers identified sensitive spectral bands for leaf nitrogen (LNC), phosphorus (LPC), and potassium content (LKC) using both CSPA and rEW-2DCOS methods. The rEW-2DCOS method, offering greater consistency and accuracy, identified spectral bands for LKC (580–820 nm, 1860–2350 nm), LNC (600–780 nm, 2100–2180 nm), and LPC (600–750 nm, 1930–2290 nm), outperforming traditional methods in nutrient inversion with improved accuracy (R²) and reduced model complexity. Additionally, a Mechanism-guided Adaptive Ensemble Learning (M-AEL) model was incorporated for nutrient estimation, adapting to different vegetation species and enhancing accuracy over conventional methods. This model demonstrated high precision in LNC, LPC, and LKC estimation, achieving an average inversion accuracy of 71%, 73%, and 71%, respectively. The study also utilized the Mantel test to explore the relationship between nutrient content and hydrology-vegetation interactions, revealing species-specific correlations with water quality and vegetation traits. These findings provide a robust method for monitoring vegetation health in karst wetlands, with potential applications in large-scale remote sensing and ecosystem management. Future research will focus on validating the model for UAV and satellite imagery to extend its applicability for broader environmental monitoring.

The results have significant implications for ecological monitoring and land management in karst wetlands. By enabling accurate, large-scale estimation of key vegetation nutrients, the method can help better assess vegetation health, predict ecosystem dynamics, and inform conservation strategies. The ability to monitor nutrient content remotely opens new opportunities for maintaining wetland ecosystems' integrity, especially in regions with limited access or frequent hydrological fluctuations. Moreover, this methodology can be applied to a variety of ecosystems beyond karst wetlands, including agriculture, forestry, and other natural habitats.

###

References

DOI

10.1016/j.plaphe.2025.100130

Original Source URl

https://doi.org/10.1016/j.plaphe.2025.100130

Funding information

This study was supported by the National Natural Science Foundation of China (Grant number 42371341), the Natural Science Foundation of Guangxi Zhuang Autonomous Region (CN) (Grant number 2025GXNSFFA069008; 2024GXNSFAA010351), the Key Laboratory of Tropical Marine Ecosystem and Bioresource, Ministry of Natural Resources (Grant Number 2023ZD02), and the Innovation Project of Guangxi Graduate Education (Grant Number YCSW2025397).

About Plant Phenomics

Plant Phenomics is dedicated to publishing novel research that will advance all aspects of plant phenotyping from the cell to the plant population levels using innovative combinations of sensor systems and data analytics. Plant Phenomics aims also to connect phenomics to other science domains, such as genomics, genetics, physiology, molecular biology, bioinformatics, statistics, mathematics, and computer sciences. Plant Phenomics should thus contribute to advance plant sciences and agriculture/forestry/horticulture by addressing key scientific challenges in the area of plant phenomics.