Estimation of Winter Wheat Stem Biomass by a Novel Two-Component and Two-Parameter Stratified Model Using Proximal Remote Sensing and Phenological Variables

作者：Weinan Chen, Guijun Yang, Meng Yang, Haikuan Feng, Heli Li, Aohua Tang, Jing Zhang, Xingang Xu, Hao Yang, Changchun Li, Zhenhong Li · 发表于：Remote Sensing · 年份：2024 · DOI：10.3390/rs16224300 · 被引用次数：11 · 研究领域：Remote Sensing in Agriculture、Remote Sensing and LiDAR Applications、Remote Sensing and Land Use

The timely and precise estimation of stem biomass is critical for monitoring the crop growing status. Optical remote sensing is limited by the penetration of sunlight into the canopy depth, and thus directly estimating winter wheat stem biomass via canopy spectra remains a difficult task. There is a stable linear relationship between the stem dry biomass (SDB) and leaf dry biomass (LDB) of winter wheat during the entire growth stage. Therefore, this study comprehensively considered remote sensing and crop phenology, as well as biomass allocation laws, to establish a novel two-component (LDB, SDB) and two-parameter (phenological variables, spectral vegetation indices) stratified model (Tc/Tp-SDB) to estimate SDB across the growth stages of winter wheat. The core of the Tc/Tp-SDB model employed phenological variables (e.g., effective accumulative temperature, EAT) to correct the SDB estimations determined from the LDB. In particular, LDB was estimated using spectral vegetation indices (e.g., red-edge chlorophyll index, CIred edge). The results revealed that the coefficient values (β0 and β1) of ordinary least squares regression (OLSR) of SDB with LDB had a strong relationship with phenological variables. These coefficient (β0 and β1) relationships were used to correct the OLSR model parameters based on the calculated phenological variables. The EAT and CIred edge were determined as the optimal parameters for predicting SDB with the novel Tc/Tp-SDB model, with r, RMSE, MAE, and ...