马尔科夫链模型在察布查尔马铃薯甲虫发生程度预测中的应用  

Application of Markov chain model in predicting the infestation level of Colorado potato beetle in Qapqal

作　　者：李超 廖江花 刘娟 孙函函 关志坚 Li Chao;Liao Jianghua;Liu Juan;Sun Hanhan;Guan Zhijian(College of Agronomy,Xinjiang Agricultural University/Key Laboratory of the Pest Monitoring and Safety Control of Crops and Forests of the Universities of Xinjiang Uygur Autonomous Region,Urumqi Xinjiang 830052,China;Qapqal County Agricultural Extension Station.Qapqal Xinjiang 835300,China)

机构地区：[1]新疆农业大学农学院/农林有害生物监测与安全防控重点实验室,新疆乌鲁木齐830052 [2]新疆维吾尔自治区察布查尔县农业技术推广站,新疆察布查尔835300

出　　处：《中国植保导刊》2022年第1期36-39,共4页China Plant Protection

基　　金：国家自然科学基金项目(31660545);农业农村部西北荒漠绿洲作物有害生物综合治理重点实验室开放基金(KFJJ201905);新疆维吾尔自治区天山英才计划。

摘　　要：以新疆察布查尔的马铃薯甲虫为研究对象,运用马尔科夫链理论,以该县1993—2021年的马铃薯甲虫田间发生程度的时间序列资料,构建了1-3阶的概率转移矩阵。察布查尔县马铃薯甲虫总体为中等发生,在早期入侵年份发生程度较重,近年来以轻度发生为主;以预报年前3年的连续发生级别预测第4年的发生状态,回检1996—2020年察布查尔县的马铃薯甲虫发生级别,历史符合率较高、达到84%。本研究为马铃薯甲虫发生程度的长期预报提供了一种准确有效的方法。We constructed the transition probability matrix at step 1, 2, and 3 based on the Markov chain theory, with the time-series data of the infestation level of Colorado potato beetle in Qapqal from 1993 to 2021. Colorado potato beetles in Qapqal generally occurred moderately, and the occurrence was relatively serious in the early years after its invasion. In recent years, this pest mainly occurred slightly. The infestation level of Colorado potato beetles in the next year can be predicted from the last three consecutive years. The infestation data of Colorado potato beetles in Qapqal County from 1996 to 2020 was verified, and prediction accuracy was relatively high, reaching 84%. This study provided an accurate and effective method for long-term prediction of the occurrence degree of Colorado potato beetles.

关 键 词：马铃薯甲虫 发生程度 预测预报 马尔科夫链 

分 类 号：S435.32[农业科学—农业昆虫与害虫防治] S431.2[农业科学—植物保护]