Monitoring the Nitrogen Nutrition Status of Rice Plants Using Spectral and Image Technologies

Abstract

Background: We aimed to investigate methods to estimate the nitrogen (N) nutrition status of rice plants using data obtained using a digital camera and a spectroradiometer. The overall aim was to compare the advantages and potential of image technology and spectral technology to monitor rice N indexes accurately, inexpensively, and in real time to optimize fertilization strategies. We conducted field trials of rice plants grown with different levels of N fertilizer in 2018 to 2019. Spectral information and images of the rice canopy were obtained at the jointing stage and the booting stage. Various image and spectral characteristic parameters were selected to construct models to estimate rice N status. Results: The determination coefficients of the models constructed using the ratio vegetation index (RVI[800,720]) and cover canopy (CC) as dependent variables were highly significant. The jointing stage was the best observation period. Among the models using spectral parameters, those constructed using RVI[800,720] to estimate rice N indexes had the highest coefficient of determination (R2) values (0.79, 0.60, and 0.61 for the models to estimate leaf area index(LAI), aboveground biomass(AGB), and plant N accumulation(PNA), respectively) (P < 0.01). Among the models using image data, those using CC to predict rice N indexes showed the highest R2 values (0.76, 0.66, and 0.73 for the models to estimate LAI, AGB, and PNA, respectively) (P < 0.01). The model using the spectral parameter RVI[800,720] had a good fit and stability in estimating leaf area index (LAI; R2 = 0.7989, root mean square error (RMSE) = 0.4427, relative RMSE (RRMSE) = 11.45), and the model using the image parameter CC had a good fit in predicting plant nitrogen accumulation (PNA; R2 = 0.6768, RMSE = 11.6925 g·m-2, RRMSE = 121.78). Conclusions: Spectral and image parameters can be used as technical parameters to estimate biomass. The spectral parameter RVI[800,720] can be used to accurately estimate LAI, and the image parameter CC can be used to accurately estimate PNA, although its stability is poor. These technical parameters can be used for fast and non-destructive monitoring of rice plant biomass.