Research On Spectral Monitoring Method Of Crop Waterlogging

With the drastic changes in the global climate,flood disasters have become one of the important disasters affecting crop yields in my country,and their impact on agriculture has severely restricted my country's economic development.Monitoring crop waterlogging has become an important area of research at home and abroad,so accurate,real-time and efficient monitoring of waterlogging is of great significance for guiding agricultural development,as well as for post-disaster crop recovery and disaster assessment.At present,the research on waterlogging mainly uses traditional methods for monitoring and early warning based on meteorological information and local actual conditions,and the research direction is focused on economic losses and environmental safety.However,there is very little research on waterlogging damage of crops,which is not limited to the planting soil.,Planting area,growth cycle and other factors,but also need to combine crop varieties,climate information and complex physiological characteristics,so there is no comprehensive and fully applicable method for the crop waterlogging monitoring system.With the development of remote sensing,remote sensing methods have made the monitoring of waterlogging in a wide range,real-time,high resolution,and accurate results.However,due to the limited time resolution,it is difficult to ensure the stability of the data,and there is a lack of accurate analysis for the identification of crop waterlogging damage.Take the artificial test field simulation of soybean and corn waterlogging experiments as an example,to study the sensitivity of waterlogging damage to phenological information and spectrum during important crop growth periods,establish crop waterlogging output estimation models and remote sensing monitoring models,and apply demonstrations in field agricultural waterlogging monitoring And evaluation is conducive to the development of agricultural informationization.Through simulated planting experiments on soybeans and corn,different levels of flooding in different growth periods of soybeans were controlled,and waterlogging damage was set to control the accumulation of water above 3cm on the soil surface.The degree of damage was distinguished by the number of days of water stress,and the number of days of waterlogging stress was set to 2.Days,4 days,6 days,8 days,10 days,and collect phenological information and spectral information after the end of the stress and 15 days after recovery.By analyzing the changes in the canopy spectrum curve and the spectral characteristic parameters of soybeans and corn before and after recovery under different waterlogging damages,the water effects are identified;for the changes in the phenological information of soybeans and corn and the spectral correlation and the sensitive band vegetation index identification effect and harvest time The degree of yield loss,combined with spectral correlation analysis to establish two crop yield estimation models,can provide a theoretical basis for remote sensing monitoring of crop phenology after waterlogging and the evaluation of yield loss,and finally establish a comprehensive crop waterlogging identification monitoring based on the sensitive vegetation index model.According to the results of changes before and after waterlogging,the sensitivity of soybean plant height,root diameter and leaf area to waterlogging stress is the strongest in the branching stage,and the water stress sensitivity of corn cob length and bald tip length is the most obvious in the filling stage.The spectral correlation of waterlogging crop phenology information is more sensitive in the invisible light near red band.It can be seen from the canopy spectral reflectance of soybeans and corn that with the increase of stress days,the spectral reflectance of visible light band becomes larger,and the spectral reflectance of near-infrared band becomes smaller;both the red edge amplitude and the red edge position can be effective in the spectral characteristic parameters Identify the waterlogging damage of two crops.in the analysis of the vegetation index to identify the waterlogging damage,the effect of the normalized vegetation index of SIPI,NDVI and GNDVI is obvious.The NDVI effect is the best in the establishment of the estimation model of the two crop yields;according to the same period with strong vegetation index changes,three types of remote sensing waterlogging monitoring models for soybeans,corn and comprehensive crops were established respectively.In addition,this paper extracts the integrated crop planting areas of Jiangjunyao Town and Nong'an County,uses integrated crop waterlogging remote sensing monitoring model for application demonstration,and combines the collection of field flood disaster information to evaluate the accuracy of the model for waterlogged crops in this area.This verifies the applicability and feasibility of remote sensing monitoring of regional crop flood disasters,provides effective basic conditions for subsequent flood disaster research,and provides a reliable theoretical basis.