Abstract:Hyperspectral remote sensing is an important technology to fulfill realtime monitoring for crop growth status based on its superior performance in acquiring vegetation canopy information rapidly and nondestructively. The objectives were to test the accuracy, reliability and adaptability of the crop growth monitoring and diagnosis 402 (CGMD-402) in crop growth monitoring and application. The experiments were carried out during 2017—2018 at Experimental Bases Farmland Irrigation Research Institute, Chinese Academy of Agricultural Sciences, Xinxiang, China. The main promoted summer maize (Zea mays L.) varieties in the north Henan plain were chosen, and nitrogen treatments included five nitrogen fertilizer application rates (0kg/hm2, 75kg/hm2, 150kg/hm2, 225kg/hm2 and 300kg/hm2 pure nitrogen, expressed as N), the leaf area index (LAI), normalized differential vegetation index (NDVI), and ratio vegetation index (RVI) of different varieties and fertilizer treatments were monitored at jointing, bellbottom, tasseling, filling and maturity stages, respectively. The NDVI and RVI were monitored by different sensors of analytical spectral devices Field-spec Pro FR-2500 spectroradiometers (ASD FR-2500) or CGMD-402, respectively. NDVI or RVI characteristics were compared by ASD FR-2500 and CGMD-402, and analyzing the quantitative relationships of vegetation index between ASD FR-2500 and CGMD-402, respectively. Then, the LAI monitoring models of corn were constructed based on CGMD-402 NDVI and CGMD-402 RVI by using correlation analysis, regression analysis and other methods. The results showed that the canopy NDVI and RVI of corn were increased with the increase of nitrogen application rate in different growth stages, and the increase amplitude were 8.20%~36.59% and 4.40%~25.16%, respectively. The correlation coefficient (R) of NDVI or RVI based on ASD FR-2500 and CGMD-402 were 0.991 and 0.985, and the determination coefficient (R2) were 0.983 and 0.969, respectively. The results indicated that there was a highly consistent of vegetation indexes based on ASD FR-2500 and CGMD-402, and the NDVI and RVI from CGMD-402 were much better than ASD FR-2500. Monitoring models based on NDVI and RVI produced better estimation for LAI, and R2 were 0.911 and 0.898. Compared the predicted value with measured value to verify reliability and applicability of monitoring model, results showed that the R2 were 0.963 and 0.954, and the relative error (RE) of the measured value and predicted value were 6.65% and 9.37%, respectively. Therefore, it was suggested that the vegetation indices of NDVI and RVI by CGMD-402 was the most suitable model for monitoring corn LAI, and there was higher prediction precision, reliability and adaptability at different growth stages, and different N rates. The results indicated that the LAI from CGMD-402 was much better than that from AccuPARLP-80. These conclusions had important implications for monitoring crop growth by CGMD-402 in the the main corn producing area.