Near-infrared spectroscopy and ensemble learning modeling for moisture detection in forest floor leaf litter

• Published in: Vibrational spectroscopy Vol. 140; p. 103841
• Main Authors: Zhu, Tao, Xing, Jian
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2025
• Subjects: Ensemble learning; Forest floor leaf litter; Moisture content detection; Near-infrared spectroscopy; Moisture content detection; Forest floor leaf litter; Ensemble learning; Near-infrared spectroscopy
• ISSN: 0924-2031
• DOI: 10.1016/j.vibspec.2025.103841

The moisture content of forest floor litter is a critical indicator for assessing forest ecosystem stability and predicting wildfire risks. Traditional near-infrared (NIR) spectroscopy methods face limitations in species applicability and model accuracy. To enhance detection generalization capability and accuracy, this study proposes a moisture content detection model optimized by differential evolution (DE) algorithm and introduces an improved triangular kernel function (ITK) for least squares support vector machine (LSSVM) regression prediction, constructing a DE-LSSVM-ITK-based litter moisture content detection model. Using forest floor litter from Quercus mongolica, Fraxinus mandshurica, and Larix gmelinii as research subjects, the model employed a ten-fold cross-validation strategy to train and ensemble 10 optimal models, with the average prediction results on the test set serving as the final output. Experimental results demonstrate that the DE-LSSVM-ITK ensemble model achieves higher prediction accuracy and robustness, making it suitable for constructing moisture content detection models for different tree species. This provides a reliable technical approach for forest ecological monitoring and fire prevention. [Display omitted] •A novel improved triangular kernel enhances LSSVM regression for leaf moisture prediction.•The DE-LSSVM-ITK model uses differential evolution for global parameter optimization.•10-fold cross-validation improves model robustness and generalization performance.•The model achieves high accuracy across broadleaf and coniferous tree species.