| Bluetongue（BT）is a infectious disease transmitted by Culicoide,mainly infecting ruminants.Outbreaks of bluetongue cause severe economic losses.Ruminants occupy a major position in Chinese livestock industry,and it has been proven that there is a medium for spreading BTV in China.Therefore,the entry of foreign Bluetongue virus（BTV）into China is likely to cause a large-scale epidemic of bluetongue.In this thesis,spatiotemporal characteristics were used to analyze the distribution of bluetongue disease.Phylogenetic tree,transmission dynamics and cellular automata were used to analyze the source of bluetongue in China and to simulate the epidemic transmission pattern.According to the key control points of the import and export trade of animals and animal products,a risk event tree was established to analyze the possibility of BTV entering China through trade.The Maxent model was used to predict the distribution of bluetongue in China,and a machine learning algorithm combined with Google Trends data to establish a French bluetongue outbreak prediction system.The main research contents and conclusions are as follows:1.Spatial and temporal characteristics of bluetongue in China and traceability of BTVThe clustering of bluetongue in China was analyzed based on the recorded data on bluetongue in China from 2010 to 2020.The positive rates of bluetongue in cattle and sheep in Guangdong（69.6%）and Yunnan（36.8%）were than the other provinces.Global autocorrelation suggested that Chinese cattle and sheep infected with bluetongue presented a clustered distribution.Combined with local spatial autocorrelation analysis and spatio-temporal scanning to determine the high-risk areas of bluetongue infection,the results showed that the high-risk areas of cattle infection were Hunan and Guangxi,and the high-risk areas of sheep infection were Sichuan,Yunnan and Guizhou and Hunan.The phylogenetic tree showed that Chinese BTV was closely related to Australia,Indonesia,Italy and France.2.Transmission dynamics and automated cellular of bluetongue in ChinaThe transmission dynamics model showed that the number of infected livestock and culicoides in China increased and then decreased.Bluetongue epidemics will persist with temperatures greater than 29°C and a basic regeneration number（R0）>1.The introduction of increased intervention nodes in the model was found to show that increasing the extermination rate of Culex midge,increasing the immunization rate of animals and the extermination of infected livestock were effective in preventing and controlling the epidemic of bluetongue.Spatio-temporal simulation of the spread of bluetongue showed a circular wave with an increase in temperature accelerating the spread of the outbreak assuming a uniform distribution of livestock and no interventions.3.Prediction of high-risk areas for bluetongue outbreaks in ChinaBased on the collected records of bluetongue infection and Culicoides distribution in China,as well as bioclimatic variables,a Maxent model was conducted to establish a model to predict the distribution of bluetongue suitability in China.The results showed that the high bioclimatic suitable areas for bluetongue infection in China were located in Xinjiang,Yunnan and Guangxi.Mean annual temperature,precipitation in the driest month and sheep density had a large effect on the model.The most suitable areas for Culicoides were located in Hubei,Sichuan and Shandong.The altitude,the precipitation of the wettest month and the average annual temperature have a great influence on the model.The composite risk map of bluetongue was obtained by superimposing the bluetongue infection and Culicoides prediction results.The high-risk areas of bluetongue in China were distributed in southern,central and some eastern parts of China,as well as parts of southwest China and southern Taiwan.4.Analysis of spatiotemporal characteristics of bluetongue in France and risk analysis and early warning of bluetongue introduced into ChinaSpatio-temporal aggregation analysis was used to study the outbreak of bluetongue outbreak data epidemic in France from 2014 to 2019,and the results showed an aggregated distribution of bluetongue outbreaks in France in 2016 and 2017,mainly in south-central France and Corsica,with the peak of bluetongue outbreaks from September to December.The standard deviation ellipse and linear direction mean results indicated that bluetongue showed a pattern of moving from the center to the east.The risk of BTV exported from bovine semen from France was assessed.Analysis showed that the average probability values of BTV-8 and other BTV serotypes in bovine semen exported from semen collection centers were 1.11×10-10 and 5.53×10-14respectively.The analysis of the possibility,consequence and hazard degree of BTV being introduced into China through French bovine semen showed that the total risk was low.The bluetongue outbreak prediction system in France was constructed by combining the Google trend data with the bluetongue symptoms as keyword searches and machine learning algorithms（http://1.15.226.101:8080/Bluetongue/）.5.Risk analysis of the introduction of Australian bluetongue into ChinaA risk event tree was used to assess the risk of BTV exported via live cattle from Australia,and the results showed that the risk of BTV infection in exported live cattle gradually decreased as vaccine coverage increased.When unvaccinated,the average probability of BTV infection in live export cattle was 3.29×10-5,which corresponds to an average of 3 infected cattle for every 105 cattle exported.Combined with the analysis of the possibility,consequence and hazard degree of BTV introduced into China through live cattle from Australia,the total risk of BTV introduced into China through live cattle was low. |
