Model for predicting the nitrogen content of rice at panicle initiation stage using data from airborne hyperspectral remote sensing

Airborne hyperspectral remote sensing was used to provide data for a general-purpose model for predicting the nitrogen content of rice at panicle initiation stage using three years of data. There were significant differences between the vegetation data which were affected by the uptake of nitrogen from the soil depending on weather conditions. Therefore, the reflectance values obtained for one year may exhibit a different trend, due to the lack of vegetation. When the partial least squares regression (PLSR) models were estimated using all combinations of the three-year data, except for the model incorporating the data from 2005, correlation coefficients (r) were greater than 0.758, and the root mean squared error (RMSE) of prediction of the full-cross validation was less than 0.876 g m−2. The accuracy of the 2003–2004–2005 model was determined using five latent variables (PCs), with r = 0.938 and RMSEP = 0.774 g m−2. There were two different patterns for the regression coefficients associated with the NIR or red-edge regions. When the 2003–2004 model was validated using the data from 2005, the prediction error of the PLSR model was 1.050 g m−2. This became 2.378 g m−2 for the 2003–2005 model using the data from 2004 and 5.061 g m−2 for the 2004–2005 model with the data from 2003. There were similarities and differences for each latent variable between the 2003–2004 model and the 2003–2004–2005 model. The 2003–2004–2005 model might be more suitable for use as a general-purpose model, because it is possible to consider and validate all of the three years data.