Research On The Remote Sensing Monitoring Methods Of Different Stresses And Senescence Of Crops

Agricultural production has been a hot topic of global concern.In recent years,influenced by climate change,COVID-19 epidemics and geopolitics,agricultural production and agricultural safety have become a hot issue of global concern.Accurate monitoring of crop stress and senescence is an important part of agriculture production,and is an important guidance for accurate use and management of pesticides(fungicides,plant growth regulators,etc.)and fertilizers(nitrogen fertilizers,etc.).Remote sensing can provide efficient,nondestructive,real-time,and large-area field observations,which provides the possibility of accurate monitoring of crop stress and senescence.This paper investigates the mechanism of crop stress and senescence occurrence and spectral response mechanism for the precise management application of different pesticides(fungicides,plant growth regulators)and fertilizers(nitrogen fertilizers),as follows: research on the quantitative identification method of wheat yellow rust based on radiative transfer model(RTM)and spore spectra,research on nitrogen diagnosis method based on vegetation indices reflecting biophysical and biochemical parameters at different fertility stage,research on cotton senescence monitoring with agronomic and remote sensing indicators using Sentinel-2 satellite data.The main findings obtained are as follows.(1)Quantitative identification methods for wheat yellow rust based on wheat biophysical and biochemical parameters data and leaf/canopy scale spectral data.Firstly,we explored the variation pattern of vegetation biophysical and biochemical parameters under different severity of yellow rust,and the spectral response pattern.Next,three new vegetation indices for yellow rust were constructed based on he above spectral response pattern.They are YROI(Yellow rust optimal index),YRII(Yellow rust identification index),YRSI(Yellow rust spores index).Finally,the accuracy and robustness of the new indices were evaluated based on the field spectral data at leaf and canopy scales,and the new indices are superior to the published indices.The research results provide support for the timely and accurate quantitative identification of wheat yellow rust and guidance for the precise use of pesticide(fungicide)against the disease in the future.(2)The optimal nitrogen diagnostic method for each fertility period was investigated under normal water and fertilizer settings and nitrogen gradient settings.Firstly,we investigated the changes of nitrogen,biophysical parameters and biochemical parameters under different fertility stages and nitrogen gradient settings.Combined with the canopy spectral data,tests were conducted to obtain the best vegetation index reflecting canopy nitrogen content at different fertility stages.The accurate monitoring of nitrogen content in wheat at the erecting-jointing stage can guide the fertilization at the jointing stage and facilitate the increase of dry matter mass at the later stage of fertility.Therefore,the selection of the best monitoring index for canopy nitrogen content was also focused on the erecting-jointing stage in the study.Carotenoid/chlorophyll ratio vegetation index(CCRI)was the best monitoring index for erecting-jointing stage with the highest estimation accuracy.Also,since field crop management varies greatly between different stresses,it is important to distinguish between yellow rust and N stress in the field.Based on a deep understanding of the spectral response mechanisms of the two stresses and the selection of the best indices,a precise discrimination between the two stresses was performed,which can be used to guide specific field crop management.(3)A method for monitoring cotton senescence based on Sentinel-2 satellite data combined with agronomic and remote sensing indicators was developed.Firstly,a new remote sensing indicator,Boll area ratio(BAR),was proposed to reflect the degree of cotton senescence.The new vegetation index models BARI(Boll area ratio index)and BORI(Boll opening rate index)were developed to reflect the lint area ratio(BAR)and the agronomic indicator boll opening rate(BOR)respectively.In tests of new indices,BARI performed best in prediction and validation of lint area ratio(BAR),and BORI had the best accuracy and robustness in estimation of BOR;Finally,the relationship between the BAR and the BOR showing an exponential increase,which is consistent with the physiological pattern of cotton senescence,was explored.The main innovative contributions of the paper are as follows.(1)Combining vegetation radiative transfer model and linear mixed model,the sensitivity of biophysical and biochemical parameters and spores in spectral response under yellow rust infestation was investigated.Three novel quantitative yellow rust identification models,YROI,YRII,and YRSI,were developed to enable accurate and effective quantitative identification of wheat yellow rust.(2)According to the findings of nitrogen and biophysical & biochemical parameters of crops under different fertility stages and different nitrogen gradients,we obtained that the carotenoid/chlorophyll ratio index(CCRI)can accurately monitor the canopy nitrogen content on the erecting-jointing stage with the best robustness.(3)To address the lack of research on remote sensing monitoring of cotton senescence,new indices BARI and BORI are proposed to monitoring cotton senescence status by quantitatively evaluating the degree of boll opening,which have higher accuracy and robustness compared with the existing indices.