Response of different varieties of maize to nitrogen stress and diagnosis of leaf nitrogen using hyperspectral data

Response of different varieties of maize to nitrogen stress and diagnosis of leaf nitrogen using hyperspectral data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Response of different varieties of maize to nitrogen stress and diagnosis of leaf nitrogen using hyperspectral data Yanli Lu, Yaru Chao, Guipei song, Caie Nie, Zhiyue Cui, Lei Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1604913/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Apr, 2023 Read the published version in Scientific Reports → Version 1 posted 8 You are reading this latest preprint version Abstract Spectral technology is theoretically effective in diagnosing N stress in maize ( Zea mays L.), but its application is affected by varietal differences. In this study, the responses to N stress, leaf N spectral diagnostic models and the differences between two maize varieties were analyzed. Variety Jiyu 5817 exhibited a greater response to different N stresses in the 12-leaf stage (V12), while Zhengdan 958 displayed a greater response in the silking stage (R1). Correlation analysis showed that the spectral bands sensitive to the leaf N content were 548–556 nm and 706–721 nm at the V12 stage in Jiyu 5817 and 760–1142 nm at the R1 stage in Zhengdan 958. Compared with those of the model established without considering the differences in varieties, the coefficient of determination (R 2 ) and root mean square error (RMSE) of the N spectral diagnosis model established according to each variety were improved by 10.6% and 29.2%, respectively. It was concluded that the V12 stage of Jiyu 5817 and the R1 stage of Zhengdan 958 were the best diagnostic stages and were more sensitive to N stress, which can further guide fertilization decision-making in precision fertilization. Nitrogen Hyperspectral Diagnostic Model Variety Difference Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Nitrogen (N) is one of the key factors in plant photosynthesis, ecosystem productivity, and leaf respiration. The low N-use efficiency (NUE) of crops has always been a concern in N management, resource savings, and environmental protection. To obtain a high yield, farmers often apply an excessive amount of N fertilizer, which has resulted in a low N utilization rate, severe losses of N fertilizer and environmental pollution. Rational fertilization is based on the following main points. First, soil nutrient analysis determines the total amount of N fertilizer during the whole growth period, but the distribution of N fertilizer in different growth stages is usually based on farmer experience [1-3] . Second, determining the N status (abundance or deficiency) through plant nutrition diagnosis informs recommendations for fertilization rates according to the plant demand at that time [4] , and third, soil testing can be combined with plant nutrition diagnosis to determine the suitable N level more accurately [5] . Real-time monitoring of plant nutrient status and then supplying fertilizer on demand is a more effective way to improve fertilizer efficiency and optimize N regulation, which is an important aspect of precision fertilization. The amount and intensity of N demand for crops vary with the growth period. Timely and effective diagnosis of plant N status and determination of an accurate fertilization time are important prerequisites for achieving the synchronization of supply and demand and improving fertilizer utilization efficiency. Gebbers and Adamchuk [6] proposed visible/near infrared (VIS/NIR) hyperspectral technology as a key technology for promoting precision agriculture development and solving food security problems. In precision agriculture, the development of hyperspectral technology has provided an effective method for fast, nondestructive, and real-time monitoring of the N status [7-11] . The spectral diagnosis of the N status is based on the correlation between the reflectance, which is sensitive to N, and the N concentration. Many spectral indices have been developed [12-15] , and the algorithm of the model has been constantly optimized [16-19] . Partial least squares regression (PLSR), one of the optimization algorithms for modeling that combines the characteristics of principal component analysis, multiple linear regression analysis and canonical correlation analysis, uses data dimension reduction, information synthesis and screening technology and extracts new comprehensive components with the best interpretation ability of the system. It has been shown to be a powerful and popular method for multiple data analyses [20-21] . Ground-based hyperspectral technology can not only provide an explanation for remote sensing data analysis on a large scale but also provide a way for farmers to predict N status in a timely manner and fertilize at the right time. Most of the previous studies focused on monitoring the crop canopy to diagnose the chlorophyll or N status by spectral techniques at different scales [22-27] . However, canopy spectral information is mixed with plant, soil, and other unknown information. Furthermore, when canopy remote sensing reflected nitrogen stress information, the plant was in a serious nitrogen stress state, and the best time window for nitrogen supplementation was missed. Agriculture in China is characterized by highly dispersed small-scale farm household management patterns, and it is more practically feasible to take the farmer’s field as a management unit. The N concentration of the leaves sensitive to spectra is also affected directly by plant N status and should reflect any N deficiency conditions. Additionally, leaf diagnosis can provide more accurate information for fertilization decision-making. There have been many studies on the determination of the sensitive reproductive period for spectral diagnosis, the determination of the sensitive leaf, and the optimization of algorithms for diagnostic models [21,27,28] . There are many models of N nutrition based on spectral data in crops, and the visible bands sensitive to chlorophyll and NIR at 680–1100 nm made greater contributions to the models. However, the N diagnosis model is limited in its application; this limitation is partly due to the problem of the model itself and mostly due to the influence of varietal characteristics [29-30] . The physiological response of maize to N deficiency is always the basis of N diagnosis using spectral data. A clear understanding of the response of different crops or their cultivars to N stress is important for improving the precision of spectral diagnosis. The objective of this research is to clarify the differences in maize cultivars' responses to different N fertilization rates, construct N diagnostic models and determine the optimal diagnostic time. This study provides a further basis for N stress diagnosis and N fertilizer precision regulation under the current pattern of farmer management in China. 2. Materials And Methods 2.1 The experimental field The experiments were conducted in 2019–2020. The experimental field was located in the international agricultural high technology industry park of the Chinese Academy of Agricultural Sciences, Wanzhuang Town, Langfang City, Hebei Province (116° 35 '16' E, 39° 35 '47' N), China, with a temperate continental climate and an annual mean temperature of 11.9 °C. The soil type was fluvo-aquic soil with a sandy soil texture, and the cropping system was a long rotation of summer maize and winter wheat with deep plowing and harrowing before sowing each season. The soil samples in the cultivated layer (0–20 cm) for chemical analysis were collected before sowing, and the basic physicochemical properties of the test soil are shown in Table 1. Table 1 Soil nutrient properties of the experimental field Year pH Organic matter (g/kg) Nitrate nitrogen (mg/kg) Ammonium nitrogen (mg/kg) Available phosphorus (mg/kg) Available potassium (g/kg) 2019 7.75 10.40 25.80 17.90 21.90 69.90 2020 8.07 10.70 33.70 13.10 23.50 51.40 2.2 Experimental design Four N application rates (0, 60, 120, and 180 kg/hm 2 , denoted N0, N1, N2, and N3, respectively), three replicates, and twenty-four experimental plots with an area of 32 m 2 each (4 m×8 m) were arranged in a randomized block design. As recommended by the ASI method, the optimum amount of fertilization were 180 kg/hm 2 for nitrogen (N), 90 kg/hm 2 for phosphorus (P 2 O 5 ) and 60 kg/hm 2 for potassium (K 2 O) respectively, phosphorus and potassium were applied at the seedling stage as base fertilizer. Two maize ( Zea mays L.) varieties, Jiyu 5817 and Zhengdan 958, were selected in this experiment, and both are popular varieties in North China. The planting density was 66667 plants/hm 2 for each variety, and weeds and pesticides were sprayed in a timely manner during the growth period. The main characteristics of the two maize varieties are shown in Table 2. Table 2 Main characteristics of the tested maize varieties Variety Plant height (cm) Ear position height (cm) Lines per ear Growth days Crude protein (%) Crude starch (%) Crude fat (%) Lysine (%) JY5817 260 105 16 103 8.68 79.93 4.11 0.28 ZD958 240 100 15 96 8.47 73.42 3.92 0.37 2.3 Sampling and yield determination The sampling periods were the 6-leaf stage (V6), 12-leaf stage (V12), flowering and silking stage (R1), filling stage (R2), waxing stage (R5) and ripening stage (R6). At the seedling stage, plants with the same growth trend were tagged and sampled in the next few stages. Before the VT stage (tasseling stage), the top fully expanded leaves of maize plants were collected, and after the VT stage, the ear leaves were taken, which are collectively referred to as key functional leaves. The yield of different treatments was calculated by the weight of grains harvested in each plot. 2.4 Determination of the N content in maize leaves The samples in paper bags were placed in an oven at 105 °C for 30 minutes for enzyme fixation; then, the temperature was set to 70 °C, and drying was continued until a constant weight was reached. The dried samples were ground and passed through a 60-mesh screen, digested by the H 2 SO 4 -H 2 O 2 method, and finally assessed by an AA3 flow injection analyzer (SEAL Analytical GmbH, Norderstedt, Germany). 2.5 Measurement of spectral reflectance The spectra of different leaves were measured with an ASD FieldSpec 3 spectrometer with a high-intensity contact probe (PANalytical, B. V, Boulder, Colorado, USA; formerly Analytical Spectral Devices). After the samples were collected, the spectra were detected with the high-intensity contact probe by clamping the leaf and avoiding veins. Five positions from the tip to the base of every leaf were measured, and 10 internal scans were made for each measurement. Then, 10 spectra were averaged into one spectrum to represent the spectra of the position; the average of 5 positions was calculated for one leaf, and, finally, the average of a leaf was calculated for a sample. 2.6 Data processing and model construction Spectral data preprocessing was performed using Viewspectpro 5.7 software and Microsoft Excel, and ANOVA was performed using IBM SPSS statistical software. Multivariate data analysis software, Unscrambler 9.7, was used to construct and validate the model, which was constructed using partial least squares (PLS1) regression and cross validation. 2.7 Statement. “Zhengdan958”, the maize (Z. mays L.) cultivar that we used in the present experiment, complied with international guidelines. We complied with the IUCN Policy Statement on Research Involving Species at risk of extinction and the Convention on the Trade in Endangered Species of Wild Fauna and Flora 3. Results and analysis 3.1 Changes in the N content and N accumulation in key functional leaves of maize under different N application rates Figure 1 shows the changes in the N content and N accumulation in key functional leaves of Jiyu 5817 during the growth period. Overall, with the decrease in the N application rate, both metrics showed a downward trend with the development of growth period. There was no significant difference in the N content or N accumulation in key functional leaves when N was applied at 120 kg/hm 2 and 180 kg/hm 2 (N2 and N3 treatments), and the recommended N application rate of 180 kg/hm 2 (N3 treatment) resulted in the highest N content and N accumulation at the ripening stage (R6). When the N application rate was less than 120 kg/hm 2 , N accumulation decreased significantly with decreasing N application. Both the N content and N accumulation showed significant variation among the different treatments at the V12 stage, which is also the most important N requirement period and the key period for topdressing. Figure 2 shows the changes in the N content and N accumulation in key functional leaves of Zhengdan 958 during the growth period, and the overall change trend was consistent with that of Jiyu 5817. There was also no significant difference in the N content in key functional leaves between the N2 and N3 treatments, and the difference in N accumulation was significant among the different N treatments at the silking stage (R1). At the ripening stage (R6), the recommended N application rate of 180 kg/hm 2 (N3) resulted in the highest N content and N accumulation. At the R5-R6 stage, except for the group without N application (N0), there was no significant difference in the N content among the other three treatment groups, and there was no significant difference in leaf N accumulation between N3 and N2; however, when the N application rate was less than 120 kg/hm 2 , the N accumulation decreased significantly with increasing N stress. Comprehensively, considering both the N content and N accumulation, the period of Zhengdan 958 sensitivity to N stress was from the V12 to R1 stages, among which R1 was the most informative because the N accumulation during this stage was the most sensitive to N stress. 3.2 Yield analysis of different maize varieties As shown in Table 3, the yield was relatively sensitive to different N stresses, N deficiency significantly reduced the yield, and the average yield of Jiyu 5817 and Zhengdan 958 reached the maximum values under the N3 treatment. Compared with that under the N3 treatment, the average yield of Jiyu 5817 under the N0, N1 and N2 treatments was decreased by 24.04%, 7.76% and 3.11%, respectively, and the N2 treatment presented no significant difference from the N3 treatment. The average yield of Zhengdan 958 under the N0, N1 and N2 treatments decreased by 22.26%, 13.91% and 7.66%, respectively, and the average yield showed a significant difference under the different N treatments. According to the yield response to the different N application rates, 180 kg/hm 2 (N3) was recommended as the optimum N application rate for the two maize varieties to achieve the highest yield. The changes in the yields of Jiyu 5817 and Zhengdan 958 were basically consistent with the change in the N in the ear leaf, especially at maturity. In Zhengdan 958, there was no difference in the N content or accumulation between N2 and N3, but the yield under N3 was significantly higher than that under N2; in Jiyu 5817, there were no significant differences between N2 and N3 in the N content, N accumulation or yield. Table 3 Yields of Jiyu 5817 and Zhengdan 958 under different N rates Treatment JY5817 (kg/hm 2 ) Standard error ZD958 (kg/hm 2 ) Standard error N0 11397 c 268 12002 d 151 N1 13839 b 260 13290 c 288 N2 14536 ab 298 14255 b 321 N3 15003 a 420 15439 a 197 Note: Different lowercase letters in the table indicate a significant difference of 0.05 ( p <0.05). 3.3 Correlations between N accumulation and spectral reflectance in key functional leaves of different maize varieties The correlation between N accumulation and spectral reflectance in key functional leaves of different maize varieties was analyzed. Fig. 3 (left) shows that the correlation between N accumulation and spectral reflectance in key functional leaves of Jiyu 5817 reached a significant correlation at 532–565 nm, 700–716 nm and 1406–1485 nm at the 6-leaf stage (V6) and at 525–576 nm, 706–721 nm, 756–955 nm, 1421–1506 nm and 2018–2398 nm at the 12-leaf stage (V12). There was a significant correlation at 705–733 nm and 785–1138 nm and an extremely significant correlation at 1397–1519 nm, 1848–1889 nm and 2000–2430 nm at the ripening stage (R6). Figure 3 (right) shows the correlation between N accumulation and spectral reflectance in key functional leaves of Zhengdan 958. At the V6 stage, there was a significant negative correlation at 508–724 nm and 1972–2100 nm and an extremely significant correlation at 509–597 nm and 697–724 nm. A significant negative correlation was observed at 760–142 nm at the silking stage (R1) and at 712–724 nm at the ripening stage. 3.1.3 Establishment of prediction models of maize leaf N content based on partial least squares regression (PLSR) The N spectral prediction models of the two varieties were established using a partial least squares regression (PLSR) method. Figure 4 shows the changes in Y-variance with the number of principal components, which was determined to be 15 for Jiyu 5817, 14 for Zhengdan 958 and 15 for the combined dataset. Figures 5 and 6 are the regression coefficient diagrams (left) and prediction evaluation diagrams (right) of the models in Jiyu 5817 (sample size n = 165) and Zhengdan 958 (sample size n = 216), respectively. In Jiyu 5817, as shown in Figure 5, the root mean square error (RMSE) and the coefficient of determination (R 2 ) of the calibration set were 0.122 and 0.935, respectively, and the RMSE and R 2 of the validation set were 0.174 and 0.860, respectively; the top 10 central wavelengths that contribute greatly to the model are 521 nm, 689 nm, 1110 nm, 1188 nm, 1323 nm, 1421 nm, 1508 nm, 1875 nm, 2100 nm and 2200 nm. In Zhengdan 958, the RMSE and R 2 of the calibration set were 0.135 and 0.883, respectively, and the RMSE and R 2 of the validation set were 0.145 and 0.878, respectively; the top 10 central wavelengths that contributed greatly to the model were 518 nm, 559 nm, 689 nm, 1420 nm, 1585 nm, 1833 nm, 1875 nm, 2020 nm, 2109 nm and 2200 nm. The data from the two varieties were mixed for further analysis. Fig. 7 shows the regression coefficient diagram and prediction evaluation diagram of the spectral prediction model of leaf N content in the key functional leaves of two maize varieties (sample size n = 381). The RMSE of the verification set model was 0.204, the R 2 was 0.794, and the number of principal components was 15. The top 10 central wavelengths were 518 nm, 559 nm, 689 nm, 1110 nm, 1420 nm, 1513 nm, 1585 nm, 1875 nm, 2103 nm and 2200 nm. Figures 4–6 show that most of the positions of the central bands with greater contributions in the three models were essentially similar. Compared with that of the integrated model of the two varieties (not considering the varietal differences), the prediction accuracy of the model with variety classification was improved; the accuracy of the calibration and validation sets of the combined varieties was lower than that of the individual varieties, but the model was also validated to be statistically accepted. Therefore, to simplify the prediction of leaf N content, a general model can be applied for different varieties. However, due to the different responses to N stress by the two varieties (3.1), the effects of varietal differences should be considered in the application of the model. 3.1.4 External test of the prediction models In Unscrambler, the above models are called to be tested using external samples. A total of 25 external samples were used for the models of Jiyu 5817 and Zhengdan 958 alone, and a total of 50 samples were used to test the integrated model. In this paper, we tested the model in three ways. First, a model of the same variety, i.e., twenty-five external samples of Jiyu 5817, was used to test the model constructed for Jiyu 5817. Second, a model for the opposite variety, i.e., the Zhengdan 958 samples, was used to test the model constructed for Jiyu 5817; in turn, the Jiyu 5817 samples were used to test the model constructed for Zhengdan 958. Third, an integrated model was tested using different samples from different varieties, i.e., samples of Jiyu 5817 or Zhengdan 958 were used to test the integrated model (Table 4). The test results show that the determination coefficient R 2 of all the evaluation results was greater than 0.803 and the relative error was less than 8.98%. The order of test accuracy from high to low was found for the model of the same variety, the integrated model validation and the model of the opposite variety. Although the determination coefficients and errors of the prediction model and its validation results were not particularly desirable, as a rapid, real-time, and nondestructive N nutrition diagnostic technique that can be applied under field conditions, it is sufficient to provide a reference basis for N regulation and management during the growing period of maize. Table 4 Test results of the prediction model based on external samples Number Sample source for testing Tested model Sample number Coefficient of determination (R 2 ) Average relative error (RE %) 1 JY5817 M1 25 0.873 5.63 2 ZD958 M2 25 0.861 6.21 3 JY517+ZD958 M3 50 0.856 6.74 4 JY5817 M2 25 0.821 8.01 5 ZD958 M1 25 0.803 8.98 Note: M1 represents the PLSR prediction model of Jiyu 5817, M2 represents the PLSR prediction model of Zhengdan 958, and M3 represents the PLSR prediction model of the two varieties combined. For example, Line 1 shows that Model 1 (M1) is tested using 25 unknown samples of Jiyu 5817, while Line 4 shows that Model 2 (M2) is tested using 25 unknown samples of Jiyu 5817. (Samples of Jiyu 5817 were used to test the Zhengdan 958 model). The explanations for the other lines are similar. 4. Conclusion and discussion 4.1 Conclusion 4.1.1 The response of leaf N and yield to N stress in different maize varieties The recommended N application rate (N3) maintained a high N content and N accumulation during the whole growth period and achieved the highest yield. There were no significant differences in the N content and N accumulation of the two varieties between 120 kg/hm 2 (N2) and 180 kg/hm 2 (N3), as well as the yield of Jiyu 5817, which indicates the potential to reduce fertilizer application on the premise of improving the fertilizer utilization rate. Both the N content and N accumulation changed more with increasing N stress at the 12-leaf stage (V12) than at the other stages in Jiyu 5817, while both changed significantly at the silking stage (R1) in Zhengdan 958. Therefore, V12 and R1 were determined to be the stages sensitive to N stress in Jiyu 5817 and Zhengdan 958, respectively, which were also the periods with the highest N requirements and the key periods for topdressing in maize. This result provides a basis for future diagnosis. 4.1.2 Evaluation of N nutrition diagnostic models for different varieties Spectral reflectance had a strong relationship with plant N status at key periods of N fertilizer regulation in maize, which was proven by correlation and regression analysis in this paper. When the partial least squares method was used to establish the calibration models, there was little difference in the prediction accuracy between the model established by individual varieties and the model established by combining the two varieties, but the former yielded a slightly higher determination coefficient (R 2 ) and lower root mean square error (RMSE) than the latter. The samples outside the model were also tested and yielded similar results. However, it is worth mentioning that the responses of different varieties to N stress are different, and the sensitive period of each variety in response to N application rates should be considered. In this study, the V12 stage in Jiyu 5817 and the R1 stage in Zhengdan 958 were most suitable for diagnosing N stress. 4.2 Discussion Monitoring the nitrogen (N) status of plants using spectral technology is the premise of precision fertilization, which is part of precision agriculture. Excessive N fertilizer input has many disadvantages, such as wasted resources, environmental pollution and the destruction of soil structure. The Chinese government is paying increasing attention to the efficient utilization of fertilizer resources and the pollution caused by chemical fertilizer. In 2015, the Ministry of Agriculture proposed the "zero increase action plan for chemical fertilizers and pesticides" ( http://www.chinanews.com/gn/2015/03-20/ 7146411.shtml ). To ensure food security and environmental friendliness, it is imperative to optimize the fertilizer utilization rate. Timely and effective diagnosis of the N nutrition status plays an important role in optimizing N management and improving utilization efficiency in precision fertilization. With the development of spectral technology, an increasing number of people are using spectrometers such as SPAD, GreenSeeker, and hyperspectral instruments to monitor crop N nutrition [31-33] , even in the field of crop breeding, providing a reference for varietal screening [34-35] . Regarding the spectral diagnosis of N nutrition, researchers have always made efforts to improve the diagnostic accuracy from various angles [36,25,27-28] . The application of spectral technology in diagnosing maize N nutrition is still affected by various factors, which has resulted in limited popularization. Regardless of the type of diagnostic technology, the physiological response of the plant/leaf to N abundance and deficiency is the basis of spectral diagnosis and cannot be ignored. N is a highly mobile element in plants, and its distribution in plants is in the form of a gradient. Ciganda et al. [37] concluded that a bell-shaped curve provided a very good fit for the vertical distribution of the chlorophyll content regardless of the crop growth stage. In theory, in the early stage of crop growth, because N is preferentially supplied to the newest leaves, the response of plants to N stress is to optimize photosynthesis by changing the vertical distribution of N to prevent the upper leaves from being stressed [27] . After entering the reproductive growth stage, the emphasis is on promoting N transport to support grain filling, and the change in the N dynamics in ear leaves is particularly important. Compared with canopy spectral monitoring, leaf diagnosis also avoids differences in variety appearance and growth to a certain extent [7,38] . In this study, the key functional leaf (the top fully expanded leaf before VT and the ear leaf after VT) was taken as the diagnostic target. There is still no clear answer regarding whether varietal differences are the dominant influencing factor in the spectral diagnosis of N nutrition in maize. Zhou et al. [29] noted that varietal differences should be considered in the research and production application of hyperspectral technology to N nutrition diagnosis in maize. Regarding differences in cropping systems, Wen et al. [11] found that the best 2-band VIs and the PLS regression based on selected FDR wavelengths provided a useful exploratory tool for estimating the LNC of maize across years, ecological areas, and unsynchronized growth stages. Lu et al. [28] proposed an N stress spectral index (NSSI) based on the ratio of the target treatment to the N-sufficient treatment, which can eliminate the influence of factors such as variety, growth period and environmental conditions. In this paper, we compared the differences between two maize varieties in response to N stress and evaluated their N prediction models. The study of varietal differences provides a basis for solving the problem of whether to consider variety as a factor in the application of the model. Based on the physiological mechanism and the spectral response of N abundance and deficiency in maize, the similarities and differences in the diagnosis of N nutrition in different varieties via spectral analysis were clarified; specifically, the N content and N accumulation in maize leaves were insufficient under N stress, and the yield was significantly affected. The response of the N content of the key functional leaves to different N levels varies with the growth period, and the determination of sensitive stages provides a basis for the spectral diagnosis of N nutrition. As shown in this paper, there is little difference between the combined model and the individual models of the two varieties, so the spectral diagnostic model of leaf N content does not take the variety factor into account. However, in practical applications, because different varieties have different periods of sensitivity to different N stresses, to ensure the accuracy of diagnosis, the most sensitive period should be selected when using the model to diagnose N stress. This study provides a basis for solving the problem of whether to consider the variety factor in the application of the model. Further research will focus on combining the phenotypic and physiological characteristics of varieties, soil fertility, and the relationship to canopy spectra, and improve the diagnostic accuracy and realize precision variable fertilizer on farmer’s field. Declarations Acknowledgments The authors are grateful for the useful comments from the anonymous reviewers and for the support of the National Natural Science Foundation of China (41371292). “Data availability” statements The datasets used and/or analysed during the current study available from the corresponding author on reasonable request. Correspondence and requests for materials should be addressed to LU Yanli or Wang Lei References Ray, P. K., Jana, A. K., Maitra, D. N., Saha, M. N., & Saha, A. R. (2000). Fertilizer prescriptions on soil test basis for jute, rice and wheat in a typic ustochrept. Journal o f t he Indian Society o f Soil Science . 48, 79-84. Solaiappan, U., Maruthi, Sankar, G. R., & Subramanian, V. (2008). Soil test-based optimal fertilizer nitrogen recommendation for sustainable sorghum (sorghum bicolor) yield in semi-arid vertic inceptisols of tamil nadu. Indian Journal of Agricultural Sciences . 78(4), 285-292. Ayushi, Srivastava, A., Singh, V. K., Sen, S., & Singh, V. (2021). Soil Test-Based Optimum Integrated Plant Nutrient Supply for Attaining Targeted Yield of Finger Millet in Mollisols of Northern India. Agricu tural Res earch , http:// doi.org/10.1007/s40003-021-00543-1 Zhang, K., Yuan, Z. F., Yang, T. C., Lu, Z. Z., Cao, Q., & Tian, Y. C. (2020). Chlorophyll meter–based nitrogen fertilizer optimization algorithm and nitrogen nutrition index for in-season fertilization of paddy rice. Agronomy Journal , 112(1), 1-13. Lee, J. H., Abawi, Y., Kang, C. S., Choi, B. R., Park, K. Y., & Lee, H. J. (2011). Field-specific variable rate fertilizer application based on rice growth diagnosis and soil testing for high quality rice production. Journal of Crop Science & Biotechnology , 14(1), 79-84. Gebbers, R., & Adamchuk, V. I. (2010). Precision Agriculture and Food Security. Science , 327, 828-831. Lu, Y. L., Li, S. K., Bai, Y. L., Carol, L. J., & Wang, J. H. (2009). Canopy spatial distribution and identification using hyperspectal data in winter wheat. Communications in Soil Science & Plant Analysis , 40(7-8), 1240-1253. Yao, X., Zhu, Y., Tian, Y., Feng, W., & Cao, W. (2010). Exploring hyperspectral bands and estimation indices for leaf nitrogen accumulation in wheat. International Journal o f Applied Earth Observation a nd Geoinformation , 12, 89-100. Stellacci, A. M., Castrignanò, A., Troccoli, A., Basso, B., & Buttafuoc, O. G. (2016). Selecting optimal hyperspectral bands to discriminate nitrogen status in durum wheat: a comparison of statistical approaches. Environmental M onitoring and A ssessment , 188(3), 188-199. Lu, Y. L., Bai, Y. L., Ma, D. L., Wang, L., & Yang, L. P. (2018). Nitrogen vertical distribution and status estimation using spectral data in maize. Communications in Soil Science and Plant Analysis , 49(5), 526-536. Wen, P. F., He, J., Ning, F., Wang, R., Zhang, Y. H., & Li J. (2019). Estimating leaf nitrogen concentration considering unsynchronized maize growth stages with canopy hyperspectral technique. Ecological Indicators , 107, 105590.1-105590.16. Curran, P.J., Dungan, J.L., & Peterson, D.L. (2001). Estimating the foliar biochemical concentration of leaves with reflectance spectrometry testing the Kokaly and Clark methodologies. Remote Sens ing of Environ ment , 76, 349–359. Sims, D.A., & Gamon, J. A. (2002). Relationships between leaf pigment content and spectral reflectance across a wide range of species, leaf structures and developmental stages. Remote Sens ing of Environ ment , 81, 337–354. Gitelson, A. A., Keydan, G. P., & Merzlyak, M. N. (2006). Three-band model for noninvasive estimation of chlorophyll, carotenoids, and anthocyanin contents in higher plant leaves. Geophysical Research Letters , 33(11), 431-433. Tian, Y.C., Yao, X., Yang, J., Cao, W. X., Hannaway, D. B., & Zhu, Y. (2011). Assessing newly developed and published vegetation indices for estimating rice leaf nitrogen concentration with ground- and space-based hyperspectral reflectance. Field Crops Res , 120, 299–310. Abdel-Rahman, E.M., Mutanga, O., Odindi, J., Adam, E., Odindo, A., & Ismail, R. (2014). A comparison of partial least squares (PLS) and sparse PLS regressions for predicting yield of Swiss chard grown under different irrigation water sources using hyperspectral data. Computers a nd Electronics i n Agriculture , 106, 11–19. Mahesh, S., Jayas, D. S., Paliwal, J., & White, N. D. G. (2015). Comparison of partial least squares regression (PLSR) and principal components regression (PCR) methods for protein and hardness predictions using the near-infrared (NIR) hyperspectral images of bulk samples of canadian wheat. Food Bioprocess Technol ogy , 8, 31-40. Liang, L., Di, L. P., Zhang, L. P., Deng, M. X., Qin, Z. H., & Zhao, S. H. (2015). Estimation of crop LAI using hyperspectral vegetation indices and a hybrid inversion method. Remote Sens ing of Environ ment , 165, 123-134. Chen, W. C., Li H., Zhang, F. Y., Xiao, W. M., Zhang, R. Q., & Chen, Z. K. (2021). Handheld short-wavelength NIR spectroscopy for rapid determination of sugars and carbohydrate in fresh juice with Sampling Error Profile Analysis. Infrared Physics & Technology , (prepublish), http://doi.org/10.1016/J.INFRARED.2021.103732. Mehmood, T., Liland, K. H., Snipen, L., & Saebo, S. (2012). A review of variable selection methods in Partial Least Squares Regression. Chemometrics and Intelligent Laboratory Systems , 118, 62-69. Li, F., Mistele, B., Hu, Y., Chen, X., & Schmidhalter, U. (2014a). Reflectance estimation of canopy nitrogen content in winter wheat using optimised hyperspectral spectral indices and partial least squares regression. European Journal o f Agronomy , 52, 198-209. Inoue, Y., Sakaiya, E., Zhu, Y., & Takahashi, W., (2012). Diagnostic mapping of canopy nitrogen content inrice based on hyperspectral measurements. Remote Sens ing of Environ ment , 126, 210–221. http://doi.org /10.1016/j.rse.2012.08.026. Miphokasap, P., Honda, K., Vaiphasa, C., Souris, M., & Nagai, M. (2012). Estimating canopy nitrogen concentration in sugarcane using field imaging spectroscopy. Remote Sensing , 4, 1651-1670. Xue, L., Cao, W., Luo, W., Dai, T., & Zhu, Y. (2004). Monitoring leaf nitrogen status in rice with canopy spectral reflectance. Agronomy Journal , 96(1), 135–142. Haboudane, D., Miller, J. R., Tremblay, N., Zarco-Tejada, P. J., & Dextraze, L. (2002). Integrated narrow-band vegetation indices for prediction of crop chlorophyll content for application to precision agriculture. Remote Sens ing of Environ ment , 81(2), 416-426. Muñoz-Huerta, R., Guevara-Gonzalez, R., Contreras-Medina, L., Pacheco, I. T., Prado-Olivarez, J., & Ocampo-Velazquez, R. (2013). A review of methods for sensing the nitrogen status in plants: advantages, disadvantages and recent advances. Sensors . 13(8), 10823-10843. Li, L., Jakli, B., Lu, P., Ren, T., Ming, J., Liu, S., & Wang, S. (2018). Assessing leaf nitrogen concentration of winter oilseed rape with canopy hyperspectral technique considering a non-uniform vertical nitrogen distribution. Industrial Crops a nd Products , 116, 1-14. Klem, K., Rajsnerová, P., Novotná, K., Mìša, P., & Křen, J. (2014). Changes in vertical distribution of spectral reflectance within spring barley canopy as an indicator of nitrogen nutrition, canopy structure and yield parameters. Agriculture (Polnohospodárstvo) , 60(2), 50-59. Zhou, L. L., Feng, H. Y., Yan, Z. M., Liu, K., & Zhou, S. L. (2010). Hyperspectral diagnosis of leaf N concentration of maize and variety difference. Transactions of the CSAE , 26(8), 195－199. (in Chinese with English abstract) Monostori, I., Árendás, T., Hoffman, B., Galiba, G., Gierczik, K., & Szira, F. 2016. Relationship between SPAD value and grain yield can be affected by cultivar, environment and soil nitrogen content in wheat. Euphytica , 211(1), 103-112. Foster A. J., Kakani V. G., & Mosali J. (2016). Estimation of bioenergy crop yield and n status by hyperspectral canopy reflectance and partial least square regression. Precision Agric ulture , 18, 192–209. Pancorbo, J. L., Camino, C., Alonso-Ayuso, M., Raya-Sereno, M. D., Gonzalez-Fernandez, J., & Gabruel, L.. (2021). Simultaneous assessment of nitrogen and water status in winter wheat using hyperspectral and thermal sensors. European Journal of Agronomy . 127(3), 126287. Halle, S., Biswal, S., Naidu, B., & Bommisetty, R. (2021). Effect of Real time nitrogen management on yield attributes of direct seeded rice by using SPAD meter and Greenseeker. [Preprint.] http://doi.org /10.20944/preprints202106.0047.v1. Frels K., Guttieri M., Joyce B., Leavitt B., & Baenziger P. S. (2018). Evaluating canopy spectral reflectance vegetation indices to estimate nitrogen use traits in hard winter wheat. Field Crops Research , 217, 82-92. Zhu, S. Q., Zhang, X. R., Zhang, K., Sun, Q. F., Wan, Y. S., & Liu, F. Z. (2018). Association Analysis of Leaf SPAD Value and SSR Markers in Peanut, Molecular Plant Breeding , 16(9), 2877-2885. Ferwerda, J. G., Skidmore, A. K., & Mutanga, O. (2005). Nitrogen detection with hyperspectral normalized ratio indices across multiple plant species. Int ernational J ournal of Remote Sens ing , 26(18), 4083–4095. Ciganda, V., Gitelson, A., & Schepers, J. (2008). Vertical profile and temporal variation of chlorophyll in maize canopy: Quantitative crop vigor indicator by means of reflectance-based techniques. Agronomy Journal , 100, 1409–1417. Lu, Y. L., Li, S K., Wang, J. H., Carol, L. J., Xie, R. Z., & Wang, Z. J. (2007). Differentiating wheat varieties with different leaf angle distributions using NDVI and canopy cover. New Zealand Journal of Agricultural Research , 50(5), 1149-1156. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 11 Apr, 2023 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Major revision 17 Oct, 2022 Reviews received at journal 10 Oct, 2022 Reviewers agreed at journal 25 Sep, 2022 Reviewers invited by journal 27 May, 2022 Editor assigned by journal 27 May, 2022 Editor invited by journal 06 May, 2022 Submission checks completed at journal 06 May, 2022 First submitted to journal 28 Apr, 2022 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-1604913","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":103938660,"identity":"78b448be-ae1d-4fab-b406-1b5dd8913629","order_by":0,"name":"Yanli Lu","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yanli","middleName":"","lastName":"Lu","suffix":""},{"id":103938661,"identity":"16933bab-e919-4ec2-8798-f3a78e3df58c","order_by":1,"name":"Yaru Chao","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yaru","middleName":"","lastName":"Chao","suffix":""},{"id":103938662,"identity":"ec9d2cd7-c1b1-4201-b94f-140a2927849b","order_by":2,"name":"Guipei song","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Guipei","middleName":"","lastName":"song","suffix":""},{"id":103938663,"identity":"664e92f2-b189-49e3-94ef-fa23f1693c3d","order_by":3,"name":"Caie Nie","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Caie","middleName":"","lastName":"Nie","suffix":""},{"id":103938664,"identity":"25f1392a-884d-4169-9e27-8edccebbcc54","order_by":4,"name":"Zhiyue Cui","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhiyue","middleName":"","lastName":"Cui","suffix":""},{"id":103938665,"identity":"65401e15-8917-4134-b808-8394d7df7db8","order_by":5,"name":"Lei Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAu0lEQVRIiWNgGAWjYBACfvbG9h8fKmyY+YnWItlz+IDkjDNp7JINxGoxuOGWIM3Zcpjf4ADRtszgMTBmbDgsbXw8eQPDj4pthLXwS/cYJBfuSDc2O/OsgLHnzG0ibJlzxuDwzDPWyWY3cgyYGduI0GJwI8ewmbeNuX7zDOK1pCUz87Y5MxtIEKsFGMjHGIGBzCwB9MtBovwCjMo2BnBUtidvfPCjgggtSCCB+KhBaCFVxygYBaNgFIwQAAAflUCCJRTlHQAAAABJRU5ErkJggg==","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2022-04-28 13:14:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1604913/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1604913/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-023-31887-z","type":"published","date":"2023-04-11T20:27:31+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":21354738,"identity":"937516a3-1571-456d-b8a7-5a3dd0d69cad","added_by":"auto","created_at":"2022-05-11 17:14:46","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":280237,"visible":true,"origin":"","legend":"\u003cp\u003eN content and accumulation in the key functional leaves of Jiyu 5817 maize\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/b08bc1f75b254d2bf46bc8d4.png"},{"id":21354081,"identity":"45def0b3-9933-40ee-a155-d68ae386807f","added_by":"auto","created_at":"2022-05-11 17:09:45","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":277551,"visible":true,"origin":"","legend":"\u003cp\u003eN content and accumulation in the key functional leaves of Zhengdan 958 maize\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/b8ec81981ed7cf80ce453f83.png"},{"id":21354740,"identity":"911f63ab-726a-460c-a735-9a523bd5d3e0","added_by":"auto","created_at":"2022-05-11 17:14:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":113859,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation coefficient between N accumulation and spectral reflectance in the key functional leaves of maize\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/ea1f6597993ff17f8e2c2591.png"},{"id":21354089,"identity":"0c96f379-cf15-454c-8f6f-9ed100690c78","added_by":"auto","created_at":"2022-05-11 17:09:46","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":45424,"visible":true,"origin":"","legend":"\u003cp\u003eChanges in the Y-variance with the number of principal components\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/3ee4b8227d99b188f75047ad.png"},{"id":21354091,"identity":"c27b64c3-7bd3-4c86-ab3c-019876a8c70f","added_by":"auto","created_at":"2022-05-11 17:09:46","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":104897,"visible":true,"origin":"","legend":"\u003cp\u003e\tPLSR regression coefficients and prediction evaluation diagram for Jiyu 5817\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003eNote: The left panel is the regression coefficient figure, and the right panel is the prediction evaluation figure.\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/03b62cd1e085517e193a1fe9.png"},{"id":21355594,"identity":"1515a4aa-53db-4edf-90d2-c68de88bdc13","added_by":"auto","created_at":"2022-05-11 17:19:46","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":97010,"visible":true,"origin":"","legend":"\u003cp\u003ePLSR regression coefficients and prediction evaluation diagram for Zhengdan 958\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003eNote: The left panel is the regression coefficient figure, and the right panel is the prediction evaluation figure.\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/f2c3a7666a32f23bbd75d484.png"},{"id":21354093,"identity":"32ce1674-bcb3-476e-b92f-d3a0bfc4d1ba","added_by":"auto","created_at":"2022-05-11 17:09:46","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":121051,"visible":true,"origin":"","legend":"\u003cp\u003ePLSR regression coefficients and prediction evaluation chart of the combined maize varieties\u003c/p\u003e\u003cp\u003eNote: The left panel is the regression coefficient figure, and the right panel is the prediction evaluation figure.\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/6a0d3f9b37316249b6ff2b5f.png"},{"id":44725931,"identity":"d23ea414-c974-418a-af0a-8ec7191de254","added_by":"auto","created_at":"2023-10-16 20:44:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1471054,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1604913/v1/8444b0f5-173b-4088-bcf1-5cb55978f154.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Response of different varieties of maize to nitrogen stress and diagnosis of leaf nitrogen using hyperspectral data","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eNitrogen (N) is one of the key factors in plant photosynthesis, ecosystem productivity, and leaf respiration. The low N-use efficiency (NUE) of crops has always been a concern in N management, resource savings, and environmental protection. To obtain a high yield, farmers often apply an excessive amount of N fertilizer, which has resulted in a low N utilization rate, severe losses of N fertilizer and environmental pollution. Rational fertilization is based on the following main points. First, soil nutrient analysis determines the total amount of N fertilizer during the whole growth period, but the distribution of N fertilizer in different growth stages is usually based on farmer experience\u003csup\u003e[1-3]\u003c/sup\u003e. Second, determining the N status (abundance or deficiency) through plant nutrition diagnosis informs recommendations for fertilization rates according to the plant demand at that time\u003csup\u003e[4]\u003c/sup\u003e, and third, soil testing can be combined with plant nutrition diagnosis to determine the suitable N level more accurately\u003csup\u003e[5]\u003c/sup\u003e. Real-time monitoring of plant nutrient status and then supplying fertilizer on demand is a more effective way to improve fertilizer efficiency and optimize N regulation, which is an important aspect of precision fertilization. The amount and intensity of N demand for crops vary with the growth period. Timely and effective diagnosis of plant N status and determination of an accurate fertilization time are important prerequisites for achieving the synchronization of supply and demand and improving fertilizer utilization efficiency. Gebbers and Adamchuk\u003csup\u003e[6]\u003c/sup\u003e proposed visible/near infrared (VIS/NIR) hyperspectral technology as a key technology for promoting precision agriculture development and solving food security problems. In precision agriculture, the development of hyperspectral technology has provided an effective method for fast, nondestructive, and real-time monitoring of the N status\u003csup\u003e[7-11]\u0026nbsp;\u003c/sup\u003e. The spectral diagnosis of the N status is based on the correlation between the reflectance, which is sensitive to N, and the N concentration. Many spectral indices have been developed \u003csup\u003e[12-15]\u003c/sup\u003e, and the algorithm of the model has been constantly optimized \u003csup\u003e[16-19]\u003c/sup\u003e. Partial least squares regression (PLSR), one of the optimization algorithms for modeling that combines the characteristics of principal component analysis, multiple linear regression analysis and canonical correlation analysis, uses data dimension reduction, information synthesis and screening technology and extracts new comprehensive components with the best interpretation ability of the system. It has been shown to be a powerful and popular method for multiple data analyses\u003csup\u003e[20-21]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eGround-based hyperspectral technology can not only provide\u0026nbsp;an\u0026nbsp;explanation for remote sensing data analysis on a large scale but also provide a way for farmers to predict N status\u0026nbsp;in a timely manner\u0026nbsp;and fertilize\u0026nbsp;at the\u0026nbsp;right time.\u0026nbsp;Most of the previous studies focused on monitoring the crop canopy to diagnose the chlorophyll or N status by spectral techniques\u0026nbsp;at different scales\u0026nbsp;\u003csup\u003e[22-27]\u003c/sup\u003e. However, canopy spectral information is mixed with plant, soil, and other unknown information. Furthermore,\u0026nbsp;when canopy remote sensing reflected nitrogen stress information, the plant was in a serious nitrogen stress state, and the best time window for nitrogen supplementation was missed.\u0026nbsp;Agriculture in China is characterized by highly dispersed small-scale farm household management patterns, and\u0026nbsp;it is more practically feasible to take the farmer\u0026rsquo;s field as a management unit.\u0026nbsp;The N concentration of the leaves sensitive to\u0026nbsp;spectra\u0026nbsp;is also\u0026nbsp;affected directly by plant N status and should reflect any N deficiency conditions.\u0026nbsp;Additionally,\u0026nbsp;leaf diagnosis can provide more accurate information for\u0026nbsp;fertilization decision-making.\u003c/p\u003e\n\u003cp\u003eThere have been many studies on the determination of the sensitive reproductive period for spectral diagnosis, the determination of the sensitive leaf, and the optimization of algorithms for diagnostic models\u003csup\u003e[21,27,28]\u003c/sup\u003e. There are many models of N nutrition based on spectral data in crops, and the visible bands sensitive to chlorophyll and NIR at 680\u0026ndash;1100 nm made greater contributions to the models. However, the N diagnosis model is limited in its application; this limitation is partly due to the problem of the model itself and mostly due to the influence of varietal characteristics\u003csup\u003e[29-30]\u003c/sup\u003e. The physiological response of maize to N deficiency is always the basis of N diagnosis using spectral data. A clear understanding of the response of different crops or their cultivars to N stress is important for improving the precision of spectral diagnosis. The objective of this research is to clarify the differences in maize cultivars\u0026apos; responses to different N fertilization rates, construct N diagnostic models and determine the optimal diagnostic time. This study provides a further basis for N stress diagnosis and N fertilizer precision regulation under the current pattern of farmer management in China.\u003c/p\u003e"},{"header":"2. Materials And Methods","content":"\u003cp\u003e\u003cstrong\u003e2.1 The experimental field\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe experiments were conducted in 2019\u0026ndash;2020. The experimental field was located in the international agricultural high technology industry park of the Chinese Academy of Agricultural Sciences, Wanzhuang Town, Langfang City, Hebei Province (116\u0026deg; 35 \u0026apos;16\u0026apos; E, 39\u0026deg; 35 \u0026apos;47\u0026apos; N), China, with a temperate continental climate and an annual mean temperature of 11.9 \u0026deg;C. The soil type was fluvo-aquic\u0026nbsp;soil with a sandy soil texture, and the cropping system was a long rotation of summer maize and winter wheat with deep plowing and harrowing before sowing each season. The soil samples in the cultivated layer (0\u0026ndash;20 cm) for chemical analysis were collected before sowing, and the basic physicochemical properties of the test soil are shown in Table 1.\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eTable 1\u0026nbsp;\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eSoil nutrient properties of the experimental field\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"102%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"6.25%\"\u003e\n \u003cp\u003eYear\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"6.25%\"\u003e\n \u003cp\u003epH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.541666666666666%\"\u003e\n \u003cp\u003eOrganic matter (g/kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"15.625%\"\u003e\n \u003cp\u003eNitrate nitrogen (mg/kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"20.833333333333332%\"\u003e\n \u003cp\u003eAmmonium nitrogen\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(mg/kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"19.791666666666668%\"\u003e\n \u003cp\u003eAvailable phosphorus (mg/kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"17.708333333333332%\"\u003e\n \u003cp\u003eAvailable potassium (g/kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"6.25%\"\u003e\n \u003cp\u003e2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"6.25%\"\u003e\n \u003cp\u003e7.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.541666666666666%\"\u003e\n \u003cp\u003e10.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"15.625%\"\u003e\n \u003cp\u003e25.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"20.833333333333332%\"\u003e\n \u003cp\u003e17.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"19.791666666666668%\"\u003e\n \u003cp\u003e21.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"17.708333333333332%\"\u003e\n \u003cp\u003e69.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"6.25%\"\u003e\n \u003cp\u003e2020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"6.25%\"\u003e\n \u003cp\u003e8.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.541666666666666%\"\u003e\n \u003cp\u003e10.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"15.625%\"\u003e\n \u003cp\u003e33.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"20.833333333333332%\"\u003e\n \u003cp\u003e13.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"19.791666666666668%\"\u003e\n \u003cp\u003e23.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"17.708333333333332%\"\u003e\n \u003cp\u003e51.40\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003e2.2 Experimental design\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFour N application rates (0, 60, 120, and 180 kg/hm\u003csup\u003e2\u003c/sup\u003e, denoted N0, N1, N2, and N3, respectively), three replicates, and twenty-four experimental plots with an area of 32 m\u003csup\u003e2\u003c/sup\u003e each (4 m\u0026times;8 m) were arranged in a randomized block design. As recommended by the ASI method, the optimum amount of fertilization were\u0026nbsp;180 kg/hm\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e\u0026nbsp;\u003c/sup\u003efor nitrogen (N),\u0026nbsp;90 kg/hm\u003csup\u003e2\u003c/sup\u003e for\u0026nbsp;phosphorus (P\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e5\u003c/sub\u003e)\u0026nbsp;and 60 kg/hm\u003csup\u003e2\u003c/sup\u003e for\u0026nbsp;potassium\u0026nbsp;(K\u003csub\u003e2\u003c/sub\u003eO) respectively, phosphorus and potassium\u0026nbsp;were applied at the seedling stage as base fertilizer. Two maize (\u003cem\u003eZea mays\u003c/em\u003e L.)\u0026nbsp;varieties, Jiyu 5817 and Zhengdan 958, were selected in this experiment, and both are popular varieties\u0026nbsp;in North China. The planting density was 66667 plants/hm\u003csup\u003e2\u003c/sup\u003e for each variety, and weeds and pesticides were sprayed in a timely manner during the growth period. The main characteristics of the two maize varieties are shown in Table 2.\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eTable 2\u0026nbsp;\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eMain characteristics of the tested maize varieties\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"9.473684210526315%\"\u003e\n \u003cp\u003eVariety\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003ePlant height (cm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003eEar position height (cm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.526315789473685%\"\u003e\n \u003cp\u003eLines per ear\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.473684210526315%\"\u003e\n \u003cp\u003eGrowth days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003eCrude protein (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"12.631578947368421%\"\u003e\n \u003cp\u003eCrude starch (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.421052631578947%\"\u003e\n \u003cp\u003eCrude fat (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.421052631578947%\"\u003e\n \u003cp\u003eLysine (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"9.473684210526315%\"\u003e\n \u003cp\u003eJY5817\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003e260\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003e105\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.526315789473685%\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.473684210526315%\"\u003e\n \u003cp\u003e103\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003e8.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"12.631578947368421%\"\u003e\n \u003cp\u003e79.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.421052631578947%\"\u003e\n \u003cp\u003e4.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.421052631578947%\"\u003e\n \u003cp\u003e0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"9.473684210526315%\"\u003e\n \u003cp\u003eZD958\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003e240\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.526315789473685%\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.473684210526315%\"\u003e\n \u003cp\u003e96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"13.68421052631579%\"\u003e\n \u003cp\u003e8.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"12.631578947368421%\"\u003e\n \u003cp\u003e73.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.421052631578947%\"\u003e\n \u003cp\u003e3.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.421052631578947%\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003e2.3 Sampling and yield determination\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe sampling periods were the 6-leaf stage (V6), 12-leaf stage (V12), flowering and silking stage (R1), filling stage (R2), waxing stage (R5) and ripening stage (R6). At the seedling stage, plants with the same growth trend were tagged and sampled in the next few stages. Before the VT stage (tasseling stage), the top fully expanded leaves of maize plants were collected, and after the VT stage, the ear leaves were taken, which are collectively referred to as key functional leaves. The yield of different treatments was calculated by the weight of grains harvested in each plot.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.4 Determination of the N content in maize leaves\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe samples in paper bags were placed in an oven at 105 \u0026deg;C for 30 minutes\u0026nbsp;for enzyme fixation; then, the temperature was set to 70 \u0026deg;C, and drying was continued until a constant weight was reached. The dried samples were ground and passed through a 60-mesh screen, digested by the H\u003csub\u003e2\u003c/sub\u003eSO\u003csub\u003e4\u003c/sub\u003e-H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u0026nbsp;\u003c/sub\u003emethod, and finally assessed by an AA3 flow injection analyzer (SEAL Analytical GmbH,\u0026nbsp;Norderstedt, Germany).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5 Measurement of spectral reflectance\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe spectra of different leaves were measured with an ASD FieldSpec 3 spectrometer with a high-intensity contact probe (PANalytical, B. V, Boulder, Colorado, USA; formerly Analytical Spectral Devices). After the samples were collected, the spectra were detected with the high-intensity contact probe by clamping the leaf and avoiding veins. Five positions from the tip to the base of every leaf were measured, and 10 internal scans were made for each measurement. Then, 10 spectra were averaged into one spectrum to represent the spectra of the position; the average of 5 positions was calculated for one leaf, and, finally, the average of a leaf was calculated for a sample.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.6 Data processing and model construction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSpectral data preprocessing was performed using Viewspectpro 5.7 software and Microsoft Excel, and ANOVA was performed using IBM SPSS statistical software. Multivariate data analysis software, Unscrambler 9.7, was used to construct and validate the model, which was constructed using partial least squares (PLS1) regression and cross validation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.7 Statement. \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026ldquo;Zhengdan958\u0026rdquo;, the maize (Z. mays L.) cultivar that we used in the present experiment, complied with international guidelines. We complied with the IUCN Policy Statement on Research Involving Species at risk of extinction and the Convention on the Trade in Endangered Species of Wild Fauna and Flora\u003c/p\u003e"},{"header":"3. Results and analysis","content":"\u003cp\u003e\u003cstrong\u003e3.1 Changes in the N content and N accumulation in key functional leaves of maize under different N application rates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 1 shows the changes in the N content and N accumulation in key functional leaves of Jiyu 5817 during the growth period. Overall, with the decrease in the N application rate, both metrics showed a downward trend with the development of growth period. There was no significant difference in the N content or N accumulation in key functional leaves when N was applied at 120 kg/hm\u003csup\u003e2\u003c/sup\u003e and 180 kg/hm\u003csup\u003e2\u003c/sup\u003e (N2 and N3 treatments), and the recommended N application rate of 180 kg/hm\u003csup\u003e2\u003c/sup\u003e (N3 treatment) resulted in the highest N content and N accumulation at the ripening stage (R6). When the N application rate was less than 120 kg/hm\u003csup\u003e2\u003c/sup\u003e, N accumulation decreased significantly with decreasing N application. Both the N content and N accumulation showed significant variation among the different treatments at the V12 stage, which is also the most important N requirement period and the key period for topdressing.\u003c/p\u003e\n\u003cp\u003eFigure 2 shows the changes in the N content and N accumulation in key functional leaves of Zhengdan 958 during the growth period, and the overall change trend was consistent with that of Jiyu 5817. There was also no significant difference in the N content in key functional leaves between the N2 and N3 treatments, and the difference in N accumulation was significant among the different N treatments at the silking stage (R1). At the ripening stage (R6), the recommended N application rate of 180 kg/hm\u003csup\u003e2\u003c/sup\u003e (N3) resulted in the highest N content and N accumulation. At the R5-R6 stage, except for the group without N application (N0), there was no significant difference in the N content among the other three treatment groups, and there was no significant difference in leaf N accumulation between N3 and N2; however, when the N application rate was less than 120 kg/hm\u003csup\u003e2\u003c/sup\u003e, the N accumulation decreased significantly with increasing N stress. Comprehensively,\u0026nbsp;considering both the N content and N accumulation, the period of Zhengdan 958 sensitivity to N stress was from the V12 to R1 stages, among which R1 was the most informative because the N accumulation during this stage was the most sensitive to N stress.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Yield analysis of different maize varieties\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs shown in Table 3, the yield was relatively sensitive to different N stresses, N deficiency significantly reduced the yield, and the average yield of Jiyu 5817 and Zhengdan 958 reached the maximum values under the N3 treatment. Compared with that under the N3 treatment, the average yield of Jiyu 5817 under the N0, N1 and N2 treatments was decreased by 24.04%, 7.76% and 3.11%, respectively, and the N2 treatment presented no significant difference from the N3 treatment. The average yield of Zhengdan 958 under the N0, N1 and N2 treatments decreased by 22.26%, 13.91% and 7.66%, respectively, and the average yield showed a significant difference under the different N treatments. According to the yield response to the different N application rates, 180 kg/hm\u003csup\u003e2\u0026nbsp;\u003c/sup\u003e(N3) was recommended as the optimum N application rate for the two maize varieties to achieve the highest yield. The changes in the yields of Jiyu 5817 and Zhengdan 958 were basically consistent with the change in the N in the ear leaf, especially at maturity. In Zhengdan 958, there was no difference in the N content or accumulation between N2 and N3, but the yield under N3 was significantly higher than that under N2; in Jiyu 5817, there were no significant differences between N2 and N3 in the N content, N accumulation or yield.\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eTable 3\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003e\u0026nbsp;Yields of Jiyu 5817 and Zhengdan 958 under different N rates\u003c/p\u003e\n\u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"14.43298969072165%\"\u003e\n \u003cp\u003eTreatment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"24.742268041237114%\"\u003e\n \u003cp\u003eJY5817 (kg/hm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003eStandard error\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.711340206185568%\"\u003e\n \u003cp\u003eZD958 (kg/hm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003eStandard error\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"14.43298969072165%\"\u003e\n \u003cp\u003eN0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"24.742268041237114%\"\u003e\n \u003cp\u003e11397 c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e268\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.711340206185568%\"\u003e\n \u003cp\u003e12002 d\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e151\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"14.43298969072165%\"\u003e\n \u003cp\u003eN1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"24.742268041237114%\"\u003e\n \u003cp\u003e13839 b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e260\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.711340206185568%\"\u003e\n \u003cp\u003e13290 c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e288\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"14.43298969072165%\"\u003e\n \u003cp\u003eN2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"24.742268041237114%\"\u003e\n \u003cp\u003e14536 ab\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e298\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.711340206185568%\"\u003e\n \u003cp\u003e14255 b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e321\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"14.43298969072165%\"\u003e\n \u003cp\u003eN3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"24.742268041237114%\"\u003e\n \u003cp\u003e15003 a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e420\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.711340206185568%\"\u003e\n \u003cp\u003e15439 a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.556701030927837%\"\u003e\n \u003cp\u003e197\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: Different lowercase letters in the table indicate a significant difference of 0.05 (\u003cem\u003ep\u003c/em\u003e\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Correlations between N accumulation and spectral reflectance in key functional leaves of different maize varieties\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe correlation between N accumulation and spectral reflectance in key functional leaves of different maize varieties was analyzed. Fig. 3 (left) shows that the correlation between N accumulation and spectral reflectance in key functional leaves of Jiyu 5817 reached a significant correlation at 532\u0026ndash;565 nm, 700\u0026ndash;716 nm and 1406\u0026ndash;1485 nm at the 6-leaf stage (V6) and at 525\u0026ndash;576 nm, 706\u0026ndash;721 nm, 756\u0026ndash;955 nm, 1421\u0026ndash;1506 nm and 2018\u0026ndash;2398 nm at the 12-leaf stage (V12). There was a significant correlation at 705\u0026ndash;733 nm and 785\u0026ndash;1138 nm and an extremely significant correlation at 1397\u0026ndash;1519 nm, 1848\u0026ndash;1889 nm and 2000\u0026ndash;2430 nm at the ripening stage (R6). Figure 3 (right) shows the correlation between N accumulation and spectral reflectance in key functional leaves of Zhengdan 958. At the V6 stage, there was a significant negative correlation at 508\u0026ndash;724 nm and 1972\u0026ndash;2100 nm and an extremely significant correlation at 509\u0026ndash;597 nm and 697\u0026ndash;724 nm. A significant negative correlation was observed at 760\u0026ndash;142 nm at the silking stage (R1) and at 712\u0026ndash;724 nm at the ripening stage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.1.3 Establishment of prediction models of maize leaf N content based on partial least squares regression (PLSR)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe N spectral prediction models of the two varieties were established using a partial least squares regression (PLSR) method. Figure 4 shows the changes in Y-variance with the number of principal components, which was determined to be 15 for Jiyu 5817, 14 for Zhengdan 958 and 15 for the combined dataset. Figures 5 and 6 are the regression coefficient diagrams (left) and prediction evaluation diagrams (right) of the models in Jiyu 5817 (sample size n = 165) and Zhengdan 958 (sample size n = 216), respectively. In Jiyu 5817, as shown in Figure 5, the root mean square error (RMSE) and the coefficient of determination (R\u003csup\u003e2\u003c/sup\u003e) of the calibration set were 0.122 and 0.935, respectively, and the RMSE and R\u003csup\u003e2\u003c/sup\u003e of the validation set were 0.174 and 0.860, respectively; the top 10 central wavelengths that contribute greatly to the model are 521 nm, 689 nm, 1110 nm, 1188 nm, 1323 nm, 1421 nm, 1508 nm, 1875 nm, 2100 nm and 2200 nm. In Zhengdan 958, the RMSE and R\u003csup\u003e2\u003c/sup\u003e of the calibration set were 0.135 and 0.883, respectively, and the RMSE and R\u003csup\u003e2\u003c/sup\u003e of the validation set were 0.145 and 0.878, respectively; the top 10 central wavelengths that contributed greatly to the model were 518 nm, 559 nm, 689 nm, 1420 nm, 1585 nm, 1833 nm, 1875 nm, 2020 nm, 2109 nm and 2200 nm.\u003c/p\u003e\n\u003cp\u003eThe data from the two varieties were mixed for further analysis. Fig. 7 shows the regression coefficient diagram and prediction evaluation diagram of the spectral prediction model of leaf N content in the key functional leaves of two maize varieties (sample size n = 381). The RMSE of the verification set model was 0.204, the R\u003csup\u003e2\u003c/sup\u003e was 0.794, and the number of principal components was 15. The top 10 central wavelengths were 518 nm, 559 nm, 689 nm, 1110 nm, 1420 nm, 1513 nm, 1585 nm, 1875 nm, 2103 nm and 2200 nm.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFigures 4\u0026ndash;6 show that most of the positions of the central bands with greater contributions in the three models were essentially similar. Compared with that of the integrated model of the two varieties (not considering the varietal differences), the prediction accuracy of the model with variety classification was improved; the accuracy of the calibration and validation sets of the combined varieties was lower than that of the individual varieties, but the model was also validated to be statistically accepted. Therefore, to simplify the prediction of leaf N content, a general model can be applied for different varieties. However, due to the different responses to N stress by the two varieties (3.1), the effects of varietal differences should be considered in the application of the model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.1.4 External test of the prediction models\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn Unscrambler, the above models are called to be tested using external samples. A total of 25 external samples were used for the models of Jiyu 5817 and Zhengdan 958 alone, and a total of 50 samples were used to test the integrated model. In this paper, we tested the model in three ways. First, a model of the same variety, i.e., twenty-five external samples of Jiyu 5817, was used to test the model constructed for Jiyu 5817. Second, a model for the opposite variety, i.e., the Zhengdan 958 samples, was used to test the model constructed for Jiyu 5817; in turn, the Jiyu 5817 samples were used to test the model constructed for Zhengdan 958. Third, an integrated model was tested using different samples from different varieties, i.e., samples of Jiyu 5817 or Zhengdan 958 were used to test the integrated model (Table 4). The test results show that the determination coefficient R\u003csup\u003e2\u003c/sup\u003e of all the evaluation results was greater than 0.803 and the relative error was less than 8.98%. The order of test accuracy from high to low was found for the model of the same variety, the integrated model validation and the model of the opposite variety. Although the determination coefficients and errors of the prediction model and its validation results were not particularly desirable, as a rapid, real-time, and nondestructive N nutrition diagnostic technique that can be applied under field conditions, it is sufficient to provide a reference basis for N regulation and management during the growing period of maize.\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eTable 4\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003e\u0026nbsp;Test results of the prediction model based on external samples\u003c/p\u003e\n\u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNumber\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSample source for testing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTested model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSample number\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCoefficient of determination (R\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAverage relative error (RE %)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eJY5817\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eM1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.873\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5.63\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eZD958\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eM2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.861\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.21\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eJY517+ZD958\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eM3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.856\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.74\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eJY5817\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eM2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.821\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e8.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eZD958\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eM1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.803\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e8.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: M1 represents the PLSR prediction model of Jiyu 5817, M2 represents the PLSR prediction model of Zhengdan 958, and M3 represents the PLSR prediction model of the two varieties combined. For example, Line 1 shows that Model 1 (M1) is tested using 25 unknown samples of Jiyu 5817, while Line 4 shows that Model 2 (M2) is tested using 25 unknown samples of Jiyu 5817. (Samples of Jiyu 5817 were used to test the Zhengdan 958 model). The explanations for the other lines are similar.\u003c/p\u003e"},{"header":"4. Conclusion and discussion","content":"\u003cp\u003e\u003cstrong\u003e4.1 Conclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.1.1 The response of leaf N and yield to N stress in different maize varieties\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe recommended N application rate (N3) maintained a high N content and N accumulation during the whole growth period and achieved the highest yield. There were no significant differences in the N content and N accumulation of the two varieties between 120\u0026nbsp;kg/hm\u003csup\u003e2\u003c/sup\u003e (N2) and 180\u0026nbsp;kg/hm\u003csup\u003e2\u003c/sup\u003e (N3), as well as the yield of Jiyu 5817, which indicates the potential to reduce fertilizer application\u0026nbsp;on the premise of improving the fertilizer utilization rate. Both the N content and N accumulation changed more with increasing N stress at the 12-leaf stage (V12) than at the other stages in Jiyu 5817, while both changed significantly at the silking stage (R1) in Zhengdan 958.\u0026nbsp;Therefore, V12 and R1 were determined to be the stages sensitive to N stress in Jiyu 5817 and Zhengdan 958, respectively,\u0026nbsp;which were also the periods with the highest N requirements and the key periods for topdressing in maize. This result provides a basis for future diagnosis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.1.2 Evaluation of N nutrition diagnostic models for different varieties\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSpectral reflectance had a strong relationship with plant N status at key periods of N fertilizer regulation in maize, which was proven by correlation and regression analysis in this paper. When the\u0026nbsp;partial least\u0026nbsp;squares\u0026nbsp;method\u0026nbsp;was used to\u0026nbsp;establish\u0026nbsp;the\u0026nbsp;calibration\u0026nbsp;models,\u0026nbsp;there was little difference in the prediction accuracy between the model established by individual varieties and the model established by combining the two varieties, but the former yielded a slightly higher determination coefficient (R\u003csup\u003e2\u003c/sup\u003e) and lower root mean square error (RMSE) than the latter. The samples outside the model were also tested and yielded similar results. However, it is worth mentioning that the responses of different varieties to N stress are different, and the sensitive period of each variety in response to N application rates should be considered. In this study, the V12 stage in Jiyu 5817 and the R1 stage in Zhengdan 958 were most suitable for diagnosing N stress.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2 Discussion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMonitoring the nitrogen (N)\u0026nbsp;status of plants using spectral technology is the premise of precision fertilization, which is part of precision agriculture.\u0026nbsp;Excessive N fertilizer input has many disadvantages, such as wasted resources, environmental pollution and the destruction of soil structure. The Chinese government is paying increasing attention to the efficient utilization of fertilizer resources and the pollution caused by chemical fertilizer. In 2015, the Ministry of Agriculture proposed the \u0026quot;zero increase action plan for chemical fertilizers and pesticides\u0026quot; (\u003ca href=\"http://www.chinanews.com/gn/2015/03-20/%207146411.shtml\"\u003ehttp://www.chinanews.com/gn/2015/03-20/ 7146411.shtml\u003c/a\u003e). To ensure food security and environmental friendliness, it is imperative to optimize the fertilizer utilization rate. Timely and effective diagnosis of the N nutrition status plays an important role in optimizing N management and improving utilization efficiency\u0026nbsp;in precision fertilization. With the development of spectral technology, an increasing number of people are using spectrometers such as SPAD, GreenSeeker, and hyperspectral instruments to monitor crop N nutrition\u003csup\u003e[31-33]\u003c/sup\u003e, even in the field of crop breeding, providing a reference for varietal screening\u003csup\u003e[34-35]\u003c/sup\u003e. Regarding the spectral diagnosis of N nutrition, researchers have always made efforts to improve the diagnostic accuracy from various angles \u003csup\u003e[36,25,27-28]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe application of\u0026nbsp;spectral technology in diagnosing\u0026nbsp;maize N nutrition is still affected by various factors, which has\u0026nbsp;resulted in limited popularization. Regardless of the type of diagnostic technology, the physiological response of the plant/leaf to N abundance and deficiency is the basis of spectral diagnosis and cannot be ignored. N is a highly mobile element in plants, and its distribution in plants is in the form of a gradient. Ciganda et al. \u003csup\u003e[37]\u003c/sup\u003e concluded that a bell-shaped curve provided a very good fit for the vertical distribution of the chlorophyll content regardless of the crop growth stage. In theory, in the early stage of crop growth, because N is preferentially supplied to the newest leaves, the response of plants to N stress is to optimize photosynthesis by changing the vertical distribution of N to prevent the upper leaves from being stressed\u003csup\u003e[27]\u003c/sup\u003e . After entering the reproductive growth stage, the emphasis is on promoting N transport to support grain filling, and the change in the N dynamics in ear leaves is particularly important. Compared with canopy spectral monitoring, leaf diagnosis also avoids differences in variety appearance and growth to a certain extent\u003csup\u003e[7,38]\u003c/sup\u003e . In this study, the key functional leaf (the top fully expanded leaf before VT and the ear leaf after VT) was taken as the diagnostic target.\u003c/p\u003e\n\u003cp\u003eThere is still\u0026nbsp;no clear answer regarding whether varietal\u0026nbsp;differences are\u0026nbsp;the dominant influencing\u0026nbsp;factor in the spectral\u0026nbsp;diagnosis of N nutrition in\u0026nbsp;maize. Zhou\u0026nbsp;et al. \u003csup\u003e[29]\u003c/sup\u003e noted that varietal differences should be considered in the research and production application of hyperspectral technology to N nutrition diagnosis in maize. Regarding differences in cropping systems, Wen et al.\u003csup\u003e\u0026nbsp;[11]\u003c/sup\u003e found that the best 2-band VIs and the PLS regression based on selected FDR wavelengths provided a useful exploratory tool for estimating the LNC of maize across years, ecological areas, and unsynchronized growth stages. Lu et al.\u003csup\u003e[28]\u003c/sup\u003e proposed an N stress spectral index (NSSI) based on the ratio of the target treatment to the N-sufficient treatment, which can eliminate the influence of factors such as variety, growth period and environmental conditions. In this paper, we compared the differences between two maize varieties in response to N stress and evaluated their N prediction models.\u003c/p\u003e\n\u003cp\u003eThe study of varietal differences provides a basis for solving the problem of whether to consider variety as a factor in the application of the model. Based on the physiological mechanism and the spectral response of N abundance and deficiency in maize, the similarities and differences in the diagnosis of N nutrition in different varieties via spectral analysis were clarified; specifically, the N content and N accumulation in maize leaves were insufficient under N stress, and the yield was significantly affected. The response of the N content of the key functional leaves to different N levels varies with the growth period, and the determination of sensitive stages provides a basis for the spectral diagnosis of N nutrition. As shown in this paper, there is little difference between the combined model and the individual models of the two varieties, so the spectral diagnostic model of leaf N content does not take the variety factor into account. However, in practical applications, because different varieties have different periods of sensitivity to different N stresses, to ensure the accuracy of diagnosis, the most sensitive period should be selected when using the model to diagnose N stress. This study provides a basis for solving the problem of whether to consider the variety factor in the application of the model. Further research will focus on combining the phenotypic and physiological characteristics of varieties, soil fertility, and the relationship to canopy spectra, and improve the diagnostic accuracy and realize precision variable fertilizer on farmer\u0026rsquo;s field.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors are grateful for the useful comments from the anonymous reviewers and for the support of the National Natural Science Foundation of China (41371292).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026ldquo;Data availability\u0026rdquo; statements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study available from the corresponding author on reasonable request. Correspondence and requests for materials should be addressed to LU Yanli or Wang Lei\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eRay,\u0026nbsp;P. K., Jana,\u0026nbsp;A. K., Maitra,\u0026nbsp;D. N., Saha,\u0026nbsp;M. N.,\u0026nbsp;\u0026amp;\u0026nbsp;Saha,\u0026nbsp;A. R.\u0026nbsp;(2000). Fertilizer prescriptions on soil test basis for jute, rice and wheat in a typic ustochrept. \u003cem\u003eJournal\u0026nbsp;\u003c/em\u003e\u003cem\u003e\u0026nbsp;o\u003c/em\u003e\u003cem\u003ef\u0026nbsp;\u003c/em\u003e\u003cem\u003et\u003c/em\u003e\u003cem\u003ehe Indian Society\u0026nbsp;\u003c/em\u003e\u003cem\u003eo\u003c/em\u003e\u003cem\u003ef Soil Science\u003c/em\u003e. 48,\u0026nbsp;79-84.\u003c/li\u003e\n \u003cli\u003eSolaiappan,\u0026nbsp;U., Maruthi,\u0026nbsp;Sankar,\u0026nbsp;G. R.,\u0026nbsp;\u0026amp;\u0026nbsp;Subramanian,\u0026nbsp;V.\u0026nbsp;(2008). Soil test-based optimal fertilizer nitrogen recommendation for sustainable sorghum (sorghum bicolor) yield in semi-arid vertic inceptisols of tamil nadu. \u003cem\u003eIndian Journal of Agricultural Sciences\u003c/em\u003e. 78(4),\u0026nbsp;285-292.\u003c/li\u003e\n \u003cli\u003eAyushi, Srivastava, A., Singh, V. K., Sen, S., \u0026amp; Singh, V.\u0026nbsp;(2021). Soil Test-Based Optimum Integrated Plant Nutrient Supply for Attaining Targeted Yield of Finger Millet in Mollisols of Northern India. \u003cem\u003eAgricu\u003c/em\u003e\u003cem\u003etural\u003c/em\u003e\u003cem\u003e\u0026nbsp;Res\u003c/em\u003e\u003cem\u003eearch\u003c/em\u003e,\u0026nbsp;http://\u003ca href=\"https://doi.org/10.1007/s40003-021-00543-1\"\u003edoi.org/10.1007/s40003-021-00543-1\u003c/a\u003e\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eZhang, K., Yuan, Z. F., Yang, T. C., Lu, Z. Z., Cao, Q.,\u0026nbsp;\u0026amp;\u0026nbsp;Tian, Y. C.\u0026nbsp;(2020). Chlorophyll meter\u0026ndash;based nitrogen fertilizer optimization algorithm and nitrogen nutrition index for in-season fertilization of paddy rice. \u003cem\u003eAgronomy Journal\u003c/em\u003e, 112(1),\u0026nbsp;1-13.\u003c/li\u003e\n \u003cli\u003eLee,\u0026nbsp;J. H., Abawi,\u0026nbsp;Y., Kang,\u0026nbsp;C. S., Choi,\u0026nbsp;B. R., Park,\u0026nbsp;K. Y., \u0026amp;\u0026nbsp;Lee,\u0026nbsp;H. J.\u0026nbsp;(2011). Field-specific variable rate fertilizer application based on rice growth diagnosis and soil testing for high quality rice production. \u003cem\u003eJournal of Crop Science \u0026amp; Biotechnology\u003c/em\u003e, 14(1),\u0026nbsp;79-84.\u003c/li\u003e\n \u003cli\u003eGebbers,\u0026nbsp;R.,\u0026nbsp;\u0026amp;\u0026nbsp;Adamchuk,\u0026nbsp;V.\u0026nbsp;I.\u0026nbsp;(2010). Precision Agriculture and Food Security. \u003cem\u003eScience\u003c/em\u003e, 327,\u0026nbsp;828-831.\u003c/li\u003e\n \u003cli\u003eLu,\u0026nbsp;Y. L., Li,\u0026nbsp;S. K., Bai,\u0026nbsp;Y. L.,\u0026nbsp;Carol, L. J.,\u0026nbsp;\u0026amp; Wang, J. H. (2009).\u0026nbsp;Canopy spatial distribution and identification using hyperspectal data in winter wheat. \u003cem\u003eCommunications in Soil Science \u0026amp; Plant Analysis\u003c/em\u003e,\u0026nbsp;40(7-8),\u0026nbsp;1240-1253.\u003c/li\u003e\n \u003cli\u003eYao,\u0026nbsp;X., Zhu,\u0026nbsp;Y., Tian,\u0026nbsp;Y., Feng,\u0026nbsp;W.,\u0026nbsp;\u0026amp;\u0026nbsp;Cao,\u0026nbsp;W.\u0026nbsp;(2010). Exploring hyperspectral bands and estimation indices for leaf nitrogen accumulation in wheat. \u003cem\u003eInternational Journal\u0026nbsp;\u003c/em\u003e\u003cem\u003eo\u003c/em\u003e\u003cem\u003ef Applied Earth Observation\u0026nbsp;\u003c/em\u003e\u003cem\u003ea\u003c/em\u003e\u003cem\u003end Geoinformation\u003c/em\u003e,\u0026nbsp;12,\u0026nbsp;89-100.\u003c/li\u003e\n \u003cli\u003eStellacci,\u0026nbsp;A. M., Castrignan\u0026ograve;,\u0026nbsp;A., Troccoli,\u0026nbsp;A., Basso,\u0026nbsp;B.,\u0026nbsp;\u0026amp;\u0026nbsp;Buttafuoc,\u0026nbsp;O. G.\u0026nbsp;(2016). Selecting optimal hyperspectral bands to discriminate nitrogen status in durum wheat: a comparison of statistical approaches. \u003cem\u003eEnvironmental\u0026nbsp;\u003c/em\u003e\u003cem\u003eM\u003c/em\u003e\u003cem\u003eonitoring and\u0026nbsp;\u003c/em\u003e\u003cem\u003eA\u003c/em\u003e\u003cem\u003essessment\u003c/em\u003e,\u0026nbsp;188(3),\u0026nbsp;188-199.\u003c/li\u003e\n \u003cli\u003eLu,\u0026nbsp;Y. L., Bai,\u0026nbsp;Y. L., Ma,\u0026nbsp;D. L., Wang,\u0026nbsp;L.,\u0026nbsp;\u0026amp;\u0026nbsp;Yang,\u0026nbsp;L. P.\u0026nbsp;(2018). Nitrogen vertical distribution and status estimation using spectral data in maize. \u003cem\u003eCommunications in Soil Science and Plant Analysis\u003c/em\u003e,\u0026nbsp;49(5),\u0026nbsp;526-536.\u003c/li\u003e\n \u003cli\u003eWen,\u0026nbsp;P. F., He,\u0026nbsp;J., Ning,\u0026nbsp;F., Wang,\u0026nbsp;R., Zhang,\u0026nbsp;Y. H.,\u0026nbsp;\u0026amp;\u0026nbsp;Li J.\u0026nbsp;(2019). Estimating leaf nitrogen concentration considering unsynchronized maize growth stages with canopy hyperspectral technique. \u003cem\u003eEcological Indicators\u003c/em\u003e, 107, 105590.1-105590.16.\u003c/li\u003e\n \u003cli\u003eCurran,\u0026nbsp;P.J., Dungan,\u0026nbsp;J.L., \u0026amp;\u0026nbsp;Peterson,\u0026nbsp;D.L.\u0026nbsp;(2001). Estimating the foliar biochemical concentration of leaves with reflectance spectrometry testing the Kokaly and Clark methodologies. \u003cem\u003eRemote Sens\u003c/em\u003e\u003cem\u003eing of\u003c/em\u003e\u003cem\u003e\u0026nbsp;Environ\u003c/em\u003e\u003cem\u003ement\u003c/em\u003e,\u0026nbsp;76,\u0026nbsp;349\u0026ndash;359.\u003c/li\u003e\n \u003cli\u003eSims,\u0026nbsp;D.A.,\u0026nbsp;\u0026amp;\u0026nbsp;Gamon,\u0026nbsp;J. A.\u0026nbsp;(2002). Relationships between leaf pigment content and spectral reflectance across a wide range of species, leaf structures and developmental stages. \u003cem\u003eRemote Sens\u003c/em\u003e\u003cem\u003eing of\u003c/em\u003e\u003cem\u003e\u0026nbsp;Environ\u003c/em\u003e\u003cem\u003ement\u003c/em\u003e,\u0026nbsp;81,\u0026nbsp;337\u0026ndash;354.\u003c/li\u003e\n \u003cli\u003eGitelson,\u0026nbsp;A. A., Keydan,\u0026nbsp;G. P., \u0026amp;\u0026nbsp;Merzlyak,\u0026nbsp;M. N.\u0026nbsp;(2006). Three-band model for noninvasive estimation of chlorophyll, carotenoids, and anthocyanin contents in higher plant leaves. \u003cem\u003eGeophysical Research Letters\u003c/em\u003e,\u0026nbsp;33(11),\u0026nbsp;431-433.\u003c/li\u003e\n \u003cli\u003eTian,\u0026nbsp;Y.C., Yao,\u0026nbsp;X., Yang,\u0026nbsp;J., Cao,\u0026nbsp;W. X., Hannaway,\u0026nbsp;D. B.,\u0026nbsp;\u0026amp;\u0026nbsp;Zhu,\u0026nbsp;Y.\u0026nbsp;(2011). Assessing newly developed and published vegetation indices for estimating rice leaf nitrogen concentration with ground- and space-based hyperspectral reflectance. \u003cem\u003eField Crops Res\u003c/em\u003e,\u0026nbsp;120,\u0026nbsp;299\u0026ndash;310.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eAbdel-Rahman, E.M., Mutanga, O., Odindi, J., Adam, E., Odindo, A., \u0026amp;\u0026nbsp;Ismail, R.\u0026nbsp;(2014). A comparison of partial least squares (PLS) and sparse PLS regressions for predicting yield of Swiss chard grown under different irrigation water sources using hyperspectral data. \u003cem\u003eComputers\u0026nbsp;\u003c/em\u003e\u003cem\u003ea\u003c/em\u003e\u003cem\u003end Electronics\u0026nbsp;\u003c/em\u003e\u003cem\u003ei\u003c/em\u003e\u003cem\u003en Agriculture\u003c/em\u003e,\u0026nbsp;106, 11\u0026ndash;19.\u003c/li\u003e\n \u003cli\u003eMahesh,\u0026nbsp;S., Jayas,\u0026nbsp;D.\u0026nbsp;S., Paliwal,\u0026nbsp;J.,\u0026nbsp;\u0026amp;\u0026nbsp;White,\u0026nbsp;N.\u0026nbsp;D.\u0026nbsp;G.\u0026nbsp;(2015). Comparison of partial least squares regression (PLSR) and principal components regression (PCR) methods for protein and hardness predictions using the near-infrared (NIR) hyperspectral images of bulk samples of canadian wheat. \u003cem\u003eFood Bioprocess Technol\u003c/em\u003e\u003cem\u003eogy\u003c/em\u003e,\u0026nbsp;8,\u0026nbsp;31-40.\u003c/li\u003e\n \u003cli\u003eLiang,\u0026nbsp;L., Di,\u0026nbsp;L. P., Zhang,\u0026nbsp;L. P., Deng,\u0026nbsp;M. X., Qin,\u0026nbsp;Z. H., \u0026amp;\u0026nbsp;Zhao,\u0026nbsp;S. H.\u0026nbsp;(2015). Estimation of crop LAI using hyperspectral vegetation indices and a hybrid inversion method. \u003cem\u003eRemote Sens\u003c/em\u003e\u003cem\u003eing of\u003c/em\u003e\u003cem\u003e\u0026nbsp;Environ\u003c/em\u003e\u003cem\u003ement\u003c/em\u003e,\u0026nbsp;165,\u0026nbsp;123-134.\u003c/li\u003e\n \u003cli\u003eChen,\u0026nbsp;W. C., Li H., Zhang,\u0026nbsp;F. Y., Xiao,\u0026nbsp;W. M., Zhang,\u0026nbsp;R. Q., \u0026amp;\u0026nbsp;Chen,\u0026nbsp;Z. K.\u0026nbsp;(2021). Handheld short-wavelength NIR spectroscopy for rapid determination of sugars and carbohydrate in fresh juice with Sampling Error Profile Analysis. \u003cem\u003eInfrared Physics \u0026amp; Technology\u003c/em\u003e, (prepublish),\u0026nbsp;http://doi.org/10.1016/J.INFRARED.2021.103732.\u003c/li\u003e\n \u003cli\u003eMehmood,\u0026nbsp;T., Liland,\u0026nbsp;K. H., Snipen,\u0026nbsp;L.,\u0026nbsp;\u0026amp;\u0026nbsp;Saebo,\u0026nbsp;S.\u0026nbsp;(2012). A review of variable selection\u0026nbsp;methods in Partial Least Squares Regression. \u003cem\u003eChemometrics and Intelligent\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003cem\u003eLaboratory Systems\u003c/em\u003e,\u0026nbsp;118,\u0026nbsp;62-69.\u003c/li\u003e\n \u003cli\u003eLi,\u0026nbsp;F., Mistele,\u0026nbsp;B., Hu,\u0026nbsp;Y., Chen,\u0026nbsp;X., \u0026amp;\u0026nbsp;Schmidhalter,\u0026nbsp;U.\u0026nbsp;(2014a). Reflectance estimation of canopy nitrogen content in winter wheat using optimised hyperspectral spectral indices and partial least squares regression. \u003cem\u003eEuropean Journal\u0026nbsp;\u003c/em\u003e\u003cem\u003eo\u003c/em\u003e\u003cem\u003ef Agronomy\u003c/em\u003e,\u0026nbsp;52,\u0026nbsp;198-209.\u003c/li\u003e\n \u003cli\u003eInoue, Y., Sakaiya, E.,\u0026nbsp;Zhu, Y.,\u0026nbsp;\u0026amp;\u0026nbsp;Takahashi,\u0026nbsp;W., (2012). Diagnostic mapping of canopy nitrogen content inrice based on hyperspectral measurements. \u003cem\u003eRemote Sens\u003c/em\u003e\u003cem\u003eing of\u003c/em\u003e\u003cem\u003e\u0026nbsp;Environ\u003c/em\u003e\u003cem\u003ement\u003c/em\u003e,\u0026nbsp;126,\u0026nbsp;210\u0026ndash;221.\u0026nbsp;http://doi.org\u0026nbsp;/10.1016/j.rse.2012.08.026.\u003c/li\u003e\n \u003cli\u003eMiphokasap,\u0026nbsp;P., Honda,\u0026nbsp;K., Vaiphasa,\u0026nbsp;C., Souris,\u0026nbsp;M.,\u0026nbsp;\u0026amp;\u0026nbsp;Nagai,\u0026nbsp;M.\u0026nbsp;(2012). Estimating canopy nitrogen concentration in sugarcane using field imaging spectroscopy.\u003cem\u003e\u0026nbsp;Remote Sensing\u003c/em\u003e,\u0026nbsp;4,\u0026nbsp;1651-1670.\u003c/li\u003e\n \u003cli\u003eXue,\u0026nbsp;L., Cao,\u0026nbsp;W., Luo,\u0026nbsp;W., Dai,\u0026nbsp;T.,\u0026nbsp;\u0026amp;\u0026nbsp;Zhu,\u0026nbsp;Y.\u0026nbsp;(2004). Monitoring leaf nitrogen status in rice with canopy spectral reflectance. \u003cem\u003eAgronomy Journal\u003c/em\u003e,\u0026nbsp;96(1),\u0026nbsp;135\u0026ndash;142.\u003c/li\u003e\n \u003cli\u003eHaboudane,\u0026nbsp;D., Miller,\u0026nbsp;J. R., Tremblay,\u0026nbsp;N., Zarco-Tejada,\u0026nbsp;P. J., \u0026amp;\u0026nbsp;Dextraze,\u0026nbsp;L.\u0026nbsp;(2002). Integrated narrow-band vegetation indices for prediction of crop chlorophyll content for application to precision agriculture. \u003cem\u003eRemote Sens\u003c/em\u003e\u003cem\u003eing of\u003c/em\u003e\u003cem\u003e\u0026nbsp;Environ\u003c/em\u003e\u003cem\u003ement\u003c/em\u003e,\u0026nbsp;81(2),\u0026nbsp;416-426.\u003c/li\u003e\n \u003cli\u003eMu\u0026ntilde;oz-Huerta,\u0026nbsp;R., Guevara-Gonzalez,\u0026nbsp;R., Contreras-Medina,\u0026nbsp;L., Pacheco,\u0026nbsp;I. T., Prado-Olivarez,\u0026nbsp;J.,\u0026nbsp;\u0026amp;\u0026nbsp;Ocampo-Velazquez,\u0026nbsp;R.\u0026nbsp;(2013). A review of methods for sensing the nitrogen status in plants: advantages, disadvantages and recent advances. \u003cem\u003eSensors\u003c/em\u003e. 13(8),\u0026nbsp;10823-10843.\u003c/li\u003e\n \u003cli\u003eLi,\u0026nbsp;L., Jakli,\u0026nbsp;B., Lu,\u0026nbsp;P., Ren,\u0026nbsp;T., Ming,\u0026nbsp;J., Liu,\u0026nbsp;S., \u0026amp;\u0026nbsp;Wang,\u0026nbsp;S.\u0026nbsp;(2018). Assessing leaf nitrogen concentration of winter oilseed rape with canopy hyperspectral technique considering a non-uniform vertical nitrogen distribution. \u003cem\u003eIndustrial Crops\u0026nbsp;\u003c/em\u003e\u003cem\u003ea\u003c/em\u003e\u003cem\u003end Products\u003c/em\u003e,\u0026nbsp;116,\u0026nbsp;1-14.\u003c/li\u003e\n \u003cli\u003eKlem,\u0026nbsp;K., Rajsnerov\u0026aacute;,\u0026nbsp;P., Novotn\u0026aacute;,\u0026nbsp;K., M\u0026igrave;\u0026scaron;a,\u0026nbsp;P., \u0026amp;\u0026nbsp;Křen,\u0026nbsp;J.\u0026nbsp;(2014). Changes in vertical distribution of spectral reflectance within spring barley canopy as an indicator of nitrogen nutrition, canopy structure and yield parameters. \u003cem\u003eAgriculture (Polnohospod\u0026aacute;rstvo)\u003c/em\u003e, 60(2),\u0026nbsp;50-59.\u003c/li\u003e\n \u003cli\u003eZhou,\u0026nbsp;L. L., Feng,\u0026nbsp;H. Y., Yan,\u0026nbsp;Z. M., Liu,\u0026nbsp;K.,\u0026nbsp;\u0026amp;\u0026nbsp;Zhou,\u0026nbsp;S. L.\u0026nbsp;(2010). Hyperspectral diagnosis of leaf N concentration of maize and variety difference. \u003cem\u003eTransactions of the CSAE\u003c/em\u003e,\u0026nbsp;26(8), 195－199. (in Chinese with English abstract)\u003c/li\u003e\n \u003cli\u003eMonostori,\u0026nbsp;I., \u0026Aacute;rend\u0026aacute;s,\u0026nbsp;T., Hoffman,\u0026nbsp;B., Galiba,\u0026nbsp;G., Gierczik,\u0026nbsp;K.,\u0026nbsp;\u0026amp;\u0026nbsp;Szira,\u0026nbsp;F. 2016. Relationship between SPAD value and grain yield can be affected by cultivar, environment and soil nitrogen content in wheat. \u003cem\u003eEuphytica\u003c/em\u003e,\u0026nbsp;211(1),\u0026nbsp;103-112.\u003c/li\u003e\n \u003cli\u003eFoster A. J., Kakani V. G., \u0026amp;\u0026nbsp;Mosali J.\u0026nbsp;(2016). Estimation of bioenergy crop yield and n status by hyperspectral canopy reflectance and partial least square regression. \u003cem\u003ePrecision Agric\u003c/em\u003e\u003cem\u003eulture\u003c/em\u003e,\u0026nbsp;18,\u0026nbsp;192\u0026ndash;209.\u003c/li\u003e\n \u003cli\u003ePancorbo,\u0026nbsp;J. L., Camino,\u0026nbsp;C., Alonso-Ayuso,\u0026nbsp;M., Raya-Sereno,\u0026nbsp;M. D., Gonzalez-Fernandez,\u0026nbsp;J.,\u0026nbsp;\u0026amp;\u0026nbsp;Gabruel,\u0026nbsp;L..\u0026nbsp;(2021). Simultaneous assessment of nitrogen and water status in winter wheat using hyperspectral and thermal sensors. \u003cem\u003eEuropean Journal of Agronomy\u003c/em\u003e. 127(3),\u0026nbsp;126287.\u003c/li\u003e\n \u003cli\u003eHalle,\u0026nbsp;S., Biswal,\u0026nbsp;S., Naidu,\u0026nbsp;B., \u0026amp;\u0026nbsp;Bommisetty,\u0026nbsp;R.\u0026nbsp;(2021). Effect of Real time nitrogen management on yield attributes of direct seeded rice by using SPAD meter and Greenseeker.\u0026nbsp;\u0026nbsp;[Preprint.]\u0026nbsp;http://doi.org\u0026nbsp;/10.20944/preprints202106.0047.v1.\u003c/li\u003e\n \u003cli\u003eFrels K., Guttieri M., Joyce B., Leavitt B., \u0026amp;\u0026nbsp;Baenziger P. S.\u0026nbsp;(2018). Evaluating canopy spectral reflectance vegetation indices to estimate nitrogen use traits in hard winter wheat. \u003cem\u003eField Crops Research\u003c/em\u003e, 217, 82-92.\u003c/li\u003e\n \u003cli\u003eZhu,\u0026nbsp;S. Q., Zhang,\u0026nbsp;X. R., Zhang,\u0026nbsp;K., Sun,\u0026nbsp;Q. F., Wan,\u0026nbsp;Y. S.,\u0026nbsp;\u0026amp;\u0026nbsp;Liu,\u0026nbsp;F. Z.\u0026nbsp;(2018). Association Analysis of Leaf SPAD Value and SSR Markers in Peanut, \u003cem\u003eMolecular Plant Breeding\u003c/em\u003e,\u0026nbsp;16(9),\u0026nbsp;2877-2885.\u003c/li\u003e\n \u003cli\u003eFerwerda,\u0026nbsp;J. G., Skidmore,\u0026nbsp;A. K., \u0026amp;\u0026nbsp;Mutanga,\u0026nbsp;O.\u0026nbsp;(2005). Nitrogen detection with hyperspectral normalized ratio indices across multiple plant species. \u003cem\u003eInt\u003c/em\u003e\u003cem\u003eernational\u003c/em\u003e\u003cem\u003e\u0026nbsp;J\u003c/em\u003e\u003cem\u003eournal\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003cem\u003eof\u0026nbsp;\u003c/em\u003e\u003cem\u003eRemote Sens\u003c/em\u003e\u003cem\u003eing\u003c/em\u003e,\u0026nbsp;26(18),\u0026nbsp;4083\u0026ndash;4095.\u003c/li\u003e\n \u003cli\u003eCiganda,\u0026nbsp;V., Gitelson,\u0026nbsp;A., \u0026amp;\u0026nbsp;Schepers,\u0026nbsp;J.\u0026nbsp;(2008). Vertical profile and temporal variation of chlorophyll in maize canopy: Quantitative crop vigor indicator by means of reflectance-based techniques. \u003cem\u003eAgronomy Journal\u003c/em\u003e,\u0026nbsp;100,\u0026nbsp;1409\u0026ndash;1417.\u003c/li\u003e\n \u003cli\u003e\u0026nbsp;Lu,\u0026nbsp;Y. L., Li,\u0026nbsp;S K., Wang,\u0026nbsp;J. H.,\u0026nbsp;Carol, L. J., Xie, R. Z.,\u0026nbsp;\u0026amp;\u0026nbsp;Wang, Z. J. (2007).\u0026nbsp;Differentiating wheat varieties with different leaf angle distributions using NDVI and canopy cover. \u003cem\u003eNew Zealand Journal of Agricultural Research\u003c/em\u003e, 50(5), 1149-1156.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Nitrogen, Hyperspectral, Diagnostic Model, Variety Difference","lastPublishedDoi":"10.21203/rs.3.rs-1604913/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1604913/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSpectral technology is theoretically effective in diagnosing N stress in maize (\u003cem\u003eZea mays\u003c/em\u003e L.), but its application is affected by varietal differences. In this study, the responses to N stress, leaf N spectral diagnostic models and the differences between two maize varieties were analyzed. Variety Jiyu 5817 exhibited a greater response to different N stresses in the 12-leaf stage (V12), while Zhengdan 958 displayed a greater response in the silking stage (R1). Correlation analysis showed that the spectral bands sensitive to the leaf N content were 548\u0026ndash;556 nm and 706\u0026ndash;721 nm at the V12 stage in Jiyu 5817 and 760\u0026ndash;1142 nm at the R1 stage in Zhengdan 958. Compared with those of the model established without considering the differences in varieties, the coefficient of determination (R\u003csup\u003e2\u003c/sup\u003e) and root mean square error (RMSE) of the N spectral diagnosis model established according to each variety were improved by 10.6% and 29.2%, respectively. It was concluded that the V12 stage of Jiyu 5817 and the R1 stage of Zhengdan 958 were the best diagnostic stages and were more sensitive to N stress, which can further guide fertilization decision-making in precision fertilization.\u003c/p\u003e","manuscriptTitle":"Response of different varieties of maize to nitrogen stress and diagnosis of leaf nitrogen using hyperspectral data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-05-11 17:09:43","doi":"10.21203/rs.3.rs-1604913/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2022-10-17T10:35:18+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2022-10-10T08:27:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"cf705d12-e991-4b1b-9dc8-2a5de53a4c05","date":"2022-09-25T15:09:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2022-05-27T08:30:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2022-05-27T08:20:24+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2022-05-06T07:35:16+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2022-05-06T07:08:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2022-04-28T13:02:52+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c9cccb7e-cea9-4eaa-8798-46834c875bed","owner":[],"postedDate":"May 11th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2023-10-16T20:38:23+00:00","versionOfRecord":{"articleIdentity":"rs-1604913","link":"https://doi.org/10.1038/s41598-023-31887-z","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2023-04-11 20:27:31","publishedOnDateReadable":"April 11th, 2023"},"versionCreatedAt":"2022-05-11 17:09:43","video":"","vorDoi":"10.1038/s41598-023-31887-z","vorDoiUrl":"https://doi.org/10.1038/s41598-023-31887-z","workflowStages":[]},"version":"v1","identity":"rs-1604913","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1604913","identity":"rs-1604913","version":["v1"]},"buildId":"WrCJVZZCHTDjtuVLN7oU0","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}