Multi-model Comparison and Optimal Scheme Selection for LAI Remote Sensing Retrieval of Tropical Rubber Plantations

作者：Shanshan JIANG, Shengpei Dai, Zhizhong Zhao, Yuhao YANG, Yi YANG · 发表于：DOAJ (DOAJ: Directory of Open Access Journals) · 年份：2026 · DOI：10.16768/j.issn.1004-874x.2026.04.008 · 研究领域：Remote Sensing in Agriculture、Remote Sensing and LiDAR Applications、Leaf Properties and Growth Measurement

【Objective】Leaf Area Index (LAI) is a key parameter for characterizing vegetation canopy structure and physiological function, with important implications for rubber plantations growth monitoring and ecological assessment. Existing remote sensing studies on rubber plantations LAI lack systematic cross-method comparisons under a unified framework. This study aimed to identify the optimal modelling approach for tropical rubber plantations LAI remote sensing retrieval.【Method】Focusing on rubber plantations on Hainan Island, five LAI retrieval models were systematically constructed and compared, including Simple Linear Regression (SLR), Multiple Linear Regression (MLR), Random Forest (RF), Support Vector Machine (SVM), and XGBoost, based on 2017 monthly Landsat 8 imagery, and 551 valid ground-measured LAI records, under the same dataset, identical input features, and a fixed random 10-fold cross-validation framework. The input feature set comprised five vegetation indices, one derived difference feature, and two cyclically-encoded monthly temporal variables.【Result】XGBoost achieved the best overall accuracy with an R2 of 0.9072, closely followed by RF with an R2 of 0.9026, both significantly outperforming SVM and linear models. Global feature importance analysis indicated that cyclically-encoded temporal variables contributed more gain than all vegetation indices. Residual diagnostics revealed a systematic underestimation tendency in the high-LAI range across all five models, ass...