Predicting Grain Protein Content in Winter Wheat Using Hyperspectral and Meteorological Factor

Currently, most GPC prediction models by remote sensing are a little mechanism and tryin to expand at interannual and regional scales. The objective of this study is to use Hierarchical Linear Model (HLM) integrating spectral indices at anthesis and meteorological data to achieve GPC prediction at interannual scales. Results suggested that (1) spectral polygon vegetation index (SPVI) showed a high significance with GPC, with correlation coefficient (r) value of 0.38. (2) The linear model by SPVI performed low accuracy, and the determination coefficient (R2) and root mean square error (RMSE) values were 0.13 and 1.73%, respectively, while GPC model by SPVI showed poor robustness at interannual. (3) The estimation of GPC using HLM model with considering environmental variations yielded higher accuracy (R2 = 0.58 and RMSE = 1.21%) than the linear model, and the R2 and RMSE values of validation were 0.51 and 1.37%, respectively. A high consistency between the predicted GPC and the measured GPC was shown at different years. Overall, these results in this study have demonstrated the potential applicability of HLM model for GPC prediction at various years.