不同光谱指数反演马铃薯叶片氮累积量的研究

Inversion of nitrogen accumulation in potato leaf with different spectral indices

As an important derivative parameter of optical spectrum, spectral index could reflect the leaf nitrogen accumulation of crops. However, the sensitive spectral index varies with different environments and crops. In order to obtain the sensitive spectral index for potato in Inner Mongolia, field experiments were conducted in Chayouzhongqi and Hangjinqi of Inner Mongolia from 2016 to 2018, and during the potato growth period, the canopy spectrum information of potato cultivars Kexin 1 and Shepody was obtained using a handheld spectrometer (SVC HR-1024i). Based on the previous spectral indices algorithm, the correlation coefficients between the leaves nitrogen accumulation of potato (LNA) and each of the 22 spectral indices were compared, and the nitrogen nutrition diagnosis models of potato at critical growth stages were established using linear and nonlinear regression analysis. The results were as follows: (1) the red edge area was the main spectral band for inverting the LNA of potato, and Vogelmann red edge index 2 (VOG2), Vogelmann red edge index 3 (VOG3) were the sensitive spectral indices for potato LNA in Inner Mongolia, which composed of 715, 720, 726, 734, and 747 nm of spectral bands. (2) At the seedling stage, tuber formation stage or whole growth stage, the quadratic regression models (R ² > 0.75) between VOG3 and LNA could estimate better the LNA of potato under different nitrogen levels using VOG3. (3) The root mean square error (RMSE) of the models was 4.04-6.69, 9.45-10.89, 9.17-13.45 kg hm ^-2, indicating the accuracy of using the models to predict potato LNA varies with potato growth stage, and it was lower at late growth stages, while it is higher for whole growth duration. In summary, the staged modeling for potato early growth period and the unified modeling for potato later growth period could accurately estimate the potato LNA, which provides a theoretical basis and method for the application of spectral indices in the nitrogen nutrition diagnosis of potato.