| Precision agriculture management refers to a modern agricultural operation system involving quantitative decision making,variable input and positioning implementation according to an uneven spatial distribution of crop yield and differences in the crop growth environment.It is an effective method to tackle agricultural non–point–source pollution,to economise on fertilizers and pesticides and to achieve negative growth in the amount of fertilizers and pesticides application.Accurate crop nitrogen nutrient information and an effective variable nitrogen application treatment are two key links in precision agriculture management.The traditional laboratory based method of measuring nitrogen content is laborious and time consuming,and it is difficult to rapidly obtain nitrogen nutrient data of large-area crops and meet the actual needs of precision agriculture management.With the development of remote sensing technology and the sensor–manufacturing industry,rapid,non-destructive monitoring of nitrogen nutrients in crops can be achieved and provide important information for precision agriculture management.Because chloroplasts in green vegetation contain 75%nitrogen and the two are closely related,the nitrogen status of crops can be estimated from their chlorophyll(Chl)content.Based on the strong correlation between red-edge spectral characteristics and chlorophyll content,the current study on crop Chl content inversion based on red–edge bands has been reported,but some problems exist such as the red-edge spectrum is influenced by the Chl content and LAI and the method to elucidate the interaction of Chl and LAI to reduce the influence of LAI and thus increase the inversion precision of leaf chlorophyll content(LCC)remains unclear.Monitoring of LCC on the regional scale is inseparable from satellite remote sensing images.However,there are relatively few satellite sensors equipped with a red-edge band,and research on LCC inversion based on satellite red-edge band data is still in its infancy;therefore,systematic research on this topic is urgently needed.Accurate crop nitrogen nutrient status is the precondition of precision crop management decisions.In addition,an appropriate variable nitrogen fertilization treatment(VNA)is also important.How to evaluate effectively the performance of VNA is very important.The traditional evaluation method is contrastive analysis of yield or other parameters under different fertilization management,which assumes the absence of differences in spatial distribution of crop growth before fertilization among different test plots.Due to the complexity of farmland systems,factors such as soil nutrients and crop growth have inherent spatial heterogeneity;therefore,further discussion on how to truly and accurately evaluate the effect of the VNA is required.Based on the actual need for precision agriculture management and the existing key problems related to accurate monitoring of crop nitrogen content and effective evaluation of VNA,this research examined winter wheat in northern China and using spectral data from different‘Ground-Airborne-Satellite’remote–sensing platforms,especially red–edge spectral characteristics,conducted a content inversion study of winter wheat leaf chlorophyll and different variables nitrogen application treatment performance evaluation.The main content of the research and conclusions of this study are as follows:(1)To address the problem of the influence of the crop canopy structure(leaf area index)on crop LCC inversion by the vegetation index method,the research analysed a large amount of simulation data generated by the PROSAIL model and measured data on field plots,mainly discussed the red-edge spectral characteristics,selected the bands sensitive to LCC but insensitive to LAI,constructed novel vegetation indices and performed winter wheat LCC monitoring.The following results were obtained:a)Four new chlorophyll vegetation indices were established,namely,the red–edge chlorophyll absorption index(RECAI),the red-edge chlorophyll absorption index/the triangle vegetation index(RECAI/TVI),the red-edge chlorophyll absorption index/the optimized soil–regulating vegetation index(RECAI/OSAVI)and the red-edge chlorophyll absorption index/the modified triangular vegetation index(RECAI/MTVI2).b)Among all indices,the newly proposed RECAI/TVI index had the best performance with respect to LCC inversion,with an increase in R~2 of above 13.09%and decrease in RMSE of above 6.22%,compared with other indices.Moreover,it had the best performance with respect to LAI resistance,with no correlation with LAI under different LAI levels.c)By analysing the influence of different field management schemes on the inversion accuracy of LCC using the RECAI/TVI index,it was determined that upright wheat varieties,a low nitrogen application rate,absence of irrigation and early sowing may lead to a reduced inversion accuracy,which was mainly because these field management schemes led to a smaller(or larger)winter wheat group.(2)To further optimise the crop LCC inversion method,this study used airborne PHI hyperspectral data;discussed the change rule of red–edge parameters in different growth periods,varieties,nitrogen application rates and irrigation volume treatments and conducted a winter wheat LCC inversion study based on the experience statistics method(based on red–edge parameters and vegetation index)and the partial least squares regression method,respectively.The results showed that the inversion models based on the RECAI/TVI index and the average of red–edge reflectivity as well as the partial least–squares inversion model predicted winter wheat LCC with high precision.The results again proved the LCC predictive ability of the RECAI/TVI index.Red–edge bands play an important role in LCC inversion in different inversion methods.In conclusion,airborne PHI hyperspectral data are suitable for inversion studi es on winter wheat LCC,further improving the crop LCC of the remote–sensing inversion method.(3)While addressing the little–studied situation of crop LCC inversion using satellite red–edge bands,this study selected three kinds of satellite data(Rapid Eye,Sentinel–2 and En MAP)with red–edge bands to explore their inversion abilities of winter wheat LCC to provide a theoretical basis for the red–edge band set in future satellite sensors.The research results showed the following:a)It is feasible for Rapid Eye,Sentinel–2 and En MAP satellite data to conduct winter wheat LCC inversion.The prediction method based on vegetation index and the partial least squares regression method both performed well in LCC estimation.But the partital least squares regression method showed better predictive accuracy than the method based on vegetation index.b)The introduction of red–edge bands in the vegetation index is helpful to improve LCC inversion accuracy,but it is not true that a higher number of red–edge bands in vegetation index indicates higher accuracy of the LCC inversion.The results showed that the vegetation index with two red–edge bands was most sensitive to LCC.c)The inversion accuracy of the vegetation index with a long–wave red–edge band(740–783 nm)was higher than that of the vegetation index with a short–wave red–edge band;but the short–wave red–edge band played an important role in partital least squares regression method then the long–wave red–edge band.(4)Considering the traditional evaluation method of the crop variable nitrogen application treatment(VNT)without considering the inherent spatial heterogeneity of crop growth and soil nutrient before fertilization,this study,using two consecutive years of winter wheat variable rate fertilization experimental data,proposed a novel evaluation method to evaluate the effectiveness of VNT with a combination of crop growth dynamics and yield.The results revealed the following:a)Through the comparative analysis of different vegetation indices(NDVI,OSAVI,TDVI,GRNDVI,MTCI and RTVI)commonly used to characterise crop growing conditions,the MTCI index was the most sensitive to winter wheat growth(leaf area index,chlorophyll content and yield)and can be used to represent its growth.This is the first time that the MTCI index was applied to study the comprehensive characterisation of the crop growth status.b)A novel method for evaluating the effectiveness of VNA based on a comprehensive consideration of growth dynamics and yield was proposed and used to evaluate the effectiveness of different VNAs in experiments conducted in 2004–2005and 2005–2006.In 2004–2005,the effectiveness of VNA based on chlorophyll content(S)was higher than that of VNA based on the crop canopy vegetation index(Y).In 2005–2006,the VNA(Z)based on the crop canopy vegetation index and growth model was found to be optimal followed by Y and S.c)Due to the complexity of the farmland ecological environment,the effect of VNA cannot be completely demonstrated in a short period.In future,more experiments are needed to verify its effectiveness. |
PDF Full Text Request