Inversion of total photosynthetic area index of oilseed rape based on a multilayer microwave scattering semiempirical model adapted to each phenological stage

Abstract

To accurately invert the canopy parameters (Total photosynthetic area index, TPAI) of oilseed rape, a microwave characteristic layered measurement experiment was designed, and a multilayer microwave scattering semiempirical model (MLMSSM) was constructed based on the radar response changes induced by the differences in the vertical structure and canopy components at different phenological stages. This model was then applied to the main oilseed rape production areas to conduct regional TPAI inversion. Microwave characteristic layered measurement experiments were performed at six-leaf, flowering, beginning ripening and fully ripening stages of oilseed rape in the laboratory of target microwave properties (LAMP). The MLMSSM was constructed based on the LAMP-measured data, corresponding to different plant structures containing two- and three-layer submodels, and the TPAI inversion model was derived based on the correlation between the MLMSSM parameters and the crop biophysical variables. Finally, regional TPAI inversion and validation were carried out in the main oilseed rape production area (Hengyang), using Sentinel-1 SAR data. Visualization results, MLMSSM parameters and TPAI inversion results based on LAMP data all revealed the occurrence of multiple scattering interactions among distinct oilseed rape structural layers and verified the effectiveness of the measurement scheme and the model. The regional validation results showed high TPAI inversion accuracy throughout entire oilseed rape growth stages, with R² = 0.78, RMSE = 1.03, and MAE = 0.74 under VV polarization, and R² = 0.84, RMSE = 0.75, and MAE = 0.52 under VH polarization. The MLMSSM was found to significantly outperform the modified water cloud model (MWCM), increasing R² by 0.05 and 0.18 under VV and VH polarization respectively, while reducing the RMSE and MAE by 0.49–0.72. These results prove the accuracy and applicability of the MLMSSM for regional TPAI inversion of oilseed rape.

Plant phenotyping relevance

油糧ナタネの群落光合成面積指数を推定するマイクロ波計測・散乱モデルを開発し、Sentinel-1 SARで地域検証しており、植物形質取得法が中心である。

Code and data availability

公開状態または取得可能な本文経路を確認できませんでした。

No evidence-backed public reproduction asset is currently recorded.