Hyperspectral Inversion Of Physicochemical And Biochemical Parameters Of Summer Maize At Tasseling Stage Under Waterlogging Stress

In order to improve the accuracy and accuracy of summer maize monitoring under waterlogging stress,hyperspectral remote sensing was used to monitor the physiological and biochemical parameters of the plant.In this study,summer maize in tasseling stage was used as the research subject,and waterlogging stress tests were carried out.Different flooding time(3 d,5 d,7 d),different recovery time(2 d,4 d,6 d)and different nitrogen application levels(0 kg/hm²,180 kg/hm²,300 kg/hm²)were set.Hyperspectral data were collected from maize leaves before and after waterlogging stress,and physiological and biochemical indexes such as superoxide dismutase(SOD),peroxidase(POD),catalase(CAT),malondialdehyde(MDA)and chlorophyll were measured,and the correlation between spectral data and physiological and biochemical indexes of maize was analyzed.The physiological and biochemical parameters estimation model of maize based on hyperspectrum was established and verified.The results can provide technical support for large-scale,efficient and nondestructive monitoring of maize disaster.The main conclusions of this study are as follows::(1)SOD and POD enzyme activity and chlorophyll content were increased with the increase of flood time present the trend of increasing after reducing first,from the first day waterlogged stress calculation,SOD the maximum value appeared in 5 d,POD maximum value appeared in the 7 d,compared with SOD showed hysteresis,chlorophyll content under low nitrogen and nitrogen levels are reached the maximum in 3 d,The chlorophyll content reached the maximum on the 5th day under high nitrogen level,and the higher nitrogen level was beneficial to the resistance to waterlogging stress.CAT activity increased first and then decreased after flooding,and also increased first and then decreased in the recovery period,showing a"double peak"shape.When the flooding time is short,the content of MDA first increases and then decreases,but when the flooding time is long enough(more than 7 days),the content of MDA continues to increase.Studies have shown that the duration of waterlogging stress is 5 days,and the high nitrogen level of the soil can effectively alleviate the damage caused by waterlogging stress on the plant body,and help the plant resist aging.(2)With the increase of flooding time,the reflectivity of maize leaves in the visible band increased gradually,but in the near infrared band decreased gradually.With the increase of recovery time after flooding,the reflectance of maize decreased in visible band and increased in near-infrared band.Under waterlogging stress,with the increase of nitrogen level,the reflectance of maize in visible light band decreased,but in near-infrared band increased.The reflectance of the original spectrum of maize leaves varies greatly in different bands,and the reflectance of the wavelength in the range of 400-620nm and 800-1100nm has great potential for judging whether maize is affected by waterlogging disaster and distinguishing the duration of waterlogging stress of maize.(3)The correlation between the physiological and biochemical indexes of maize leaves and the original spectrum,the first-order differential spectrum,the"trilateral"parameters and the vegetation index were established respectively.The spectral characteristic parameters were extracted and the regression model was established.The higher determination coefficient R2 of the estimation model was selected as the monitoring model for the physiological and biochemical indexes of maize for verification.The results are as follows:NDVI[459,609]was selected as spectral characteristic parameter to build maize SOD monitoring model,and the regression equation was y=2535.5x+414.81(R²=0.2994);DVI[500,647]was selected as the spectral characteristic parameter to construct a corn POD monitoring model,and the regression equation was y=28209e^17.815x(R²=0.4693);DVI[459,500]was selected as the spectral characteristic parameter to construct maize CAT monitoring model,and the regression equation was y=-198.5ln(x)-365.92(R²=0.4018);DVI[390,642]was selected as the spectral characteristic parameter to construct the maize chlorophyll monitoring model,and the regression equation was y=2.8292e^-14.82x(R²=0.5105).