Studies On Water Monitoring And Diagnosis Models Of Greenhouse Tomato Based On Canopy Spectrum

As an important facility vegetable in China,greenhouse tomato is rich in nutrition and is deeply loved by consumers.Water has a great impact on the growth,development and yield of tomato.Severe water shortage will lead to short plants,curly leaves,small and light fruits.Too much water will lead to poor root respiration,fruit cracking,rotten fruit and other symptoms.Irrigation is the only way for greenhouse vegetables to obtain water,so the control of irrigation amount is very important.In the process of tomato growth,plant water should be monitored and diagnosed in real time,so as to adjust irrigation measures in time.The purpose of this study is to realize the rapid and nondestructive monitoring of tomato plant water content by modeling the canopy spectral analysis of protected tomato,and to diagnose the plant water status by constructing the critical water content model and water balance index.In this study,two two-year field experiments of two different types of tomato varieties,two nitrogen application levels and three water levels were designed.Plant height,leaf area,aboveground dry matter quality and plant water content were used as indicators to analyze and compare the growth physiological dynamics of tomato.After preprocessing the canopy spectrum,screening the characteristic bands and establishing the vegetation index,the partial least squares regression water monitoring model was constructed.The above ground dry matter mass of plants at each water level is analyzed by variance,the critical water content model of plants is constructed,and the water balance index model is constructed by the ratio of actual water content to critical water content,so as to achieve the goal of water diagnosis.In general,the main results were as follow:1.The overall plant growth showed an upward trend first,and then a downward trend,reaching the peak at the fruit expansion stage.Different varieties had the greatest impact on the growth of tomato.The impact of nitrogen was mainly reflected in the stronger plant growth,higher plant height and larger leaf area index under high nitrogen conditions.The plant water content decreased gradually before the fruit expansion stage of the first fruit branch.With the growth and expansion of the fruit,the plant water content gradually increased.The water content of C2（infinite growth type）was higher than that of Cl（limited growth type）,and the water content of N2（high nitrogen）treatment was higher than that of N1（low nitrogen）.2.There was a correlation between canopy spectral reflectance and plant water content.Under the same variety and nitrogen application rate,the spectral reflectance increased with the increase of water content.Under the same variety and irrigation rate,the spectral reflectance increases with the increase of nitrogen application rate.The difference of reflectance in near-infrared band is generally greater than that in visible band.3.Compared five pretreatment methods,two feature band extraction methods and three vegetation indexes on the accuracy of water monitoring model.Selected the best combination to construct partial least squares regression model SG-SPA-（RVI,NDVI）-PLSR,with R2 reaching 0.7772,which provides theoretical support for plant water nondestructive monitoring.4.By means of variance analysis of aboveground dry matter mass,the water limited group and non limited group of aboveground dry matter mass were divided,the concepts of critical moisture content and water balance index were put forward,the critical moisture content model and water diagnosis model were constructed.The accuracy of the diagnosis model was 72.97%,which lay a foundation for the extension research of crop water diagnosis based on water balance index and proved the feasibility of this method.