Monitoring And Early Warning Of Wheat Head Blight In Rice-wheat Rotation Area

Wheat is one of the most important food crops in China,and its output and quality are directly related to the safety of food production and food safety in our country.In recent years,due to climate change,changes in cultivation systems and farming patterns,wheat head blight has been increasing year by year.In pandemic years,wheat production can be drastically reduced.In particular,the multiple toxins produced in diseased grains seriously threaten the health of humans and animals and affected food safety.Previous studies showed that the prediction accuracy of the automatic monitoring and early warning system for wheat scab in the corn-wheat rotation area was high,while the prediction accuracy for the rice-wheat rotation area was not high and unstable,mainly because the prediction model in the system was based on the relevant data in the corn-wheat rotation area.Therefore,in the evaluation of different crop rotation accuracy of wheat gibberellic disease monitoring and early warning system,to enhance the prediction accuracy of wheat gibberellic disease monitoring and early warning system in rice-wheat rotation area,this paper made wheat scab in Taicang and Zhangjiagang of Jiangsu province as the research object,to research and develop the prediction model and the prediction accuracy of evaluation work,has made the following main research results:1.The prediction accuracy of automatic monitoring and early warning system for wheat scab was evaluated.The incidence of wheat scab was predicted by automatic monitoring and early warning system in Shaanxi,Henan,Anhui,Jiangsu and Hubei provinces.The actual head rate of wheat scab was investigated in the middle of May at the monitoring sites,and the prediction accuracy was evaluated by the maximum error reference method of Xiaoyueyan.The results showed that the prediction accuracy of wheat scab in the corn-wheat rotation area was 100%in Shaanxi and Henan provinces in 2019 and 2020,while that in the rice-wheat rotation area was only 47.5%in Anhui,Hubei and Jiangsu provinces.2.The relationship between rice stump density and rice stump carrying rate and head blight diseased ear rate was studied.Set up test sites in Northern Jiangsu(Hongze,Jiangsu),Middle Jiangsu(Jiangyan,Jiangsu),and Southern Jiangsu(Zhangjiagang,Jiangsu)to carry out field simulation experiments.Five-point sampling was used to investigate the various treatments during the wheat heading period and wax maturity period.The carrier rate of rice stumps and wheat head blight diseased ears.The results showed that the diseased ear rate was the highest in southern Jiangsu,followed by middle Jiangsu,and the lowest in northern Jiangsu.The diseased ear rates of scab in southern Jiangsu,middle Jiangsu,and northern Jiangsu are 14.80%～26.13%,5.33%～9.47%,and 0.93%～3.47%,respectively.The density of rice stumps where the peak diseased ear rate occurs is 20 clusters/m2,10 clusters/m2 and 20 clusters/m2,respectively.3.Established a prediction model for wheat head blight in the rice-wheat rotation area.Based on data from Taicang City(1990-2020)and Zhangjiagang City(2005-2020)in Jiangsu Province,such as relevant meteorological factors,field rice stump carrying rate,and scab disease ear rate,three methods which is multiple linear regression and GM gray prediction and BP neural network are used,established the prediction model of the diseased ear rate of wheat head blight in the rice-wheat rotation area respectively.The prediction accuracy of wheat head blight in Taicang City is 75%,50%and 100%,and the relative errors are respectively 6.08%,14.17%and 0.33%;the prediction accuracy of wheat head blight in Zhangjiagang City is 100%,75%and 100%,and the relative error are 4.19%,4.65%and 0.13%respectively.The BP neural network model has the highest prediction accuracy and the smallest relative error.It can be used as a built-in model of the wheat head blight monitoring and early warning system in the rice-wheat rotation area.