水稻纹枯病流行的时间动态模拟分析  

Simulation Analysis of Seasonal Development Dynamics of Rice Sheath Blight (Rhizoctonia solani)

作　　者：李成文[1] 谭万忠[1] 欧增奇[2] 张卫 殷利利[1] 王志坤[1] 周冠军[1] 

机构地区：[1]西南大学 植物保护学院,重庆400716 [2]西南大学 数学学院,重庆400715 [3]重庆市农业技术推广总站,重庆400010

出　　处：《西南师范大学学报（自然科学版）》2007年第3期85-89,共5页Journal of Southwest China Normal University(Natural Science Edition)

摘　　要：通过田间小区试验获得水稻生长季节中纹枯病(Rhizoctonia solani)相对严重度(x)随时间(t)而变化的动态消长曲线,用Epitimulator软件计算出单分子、逻辑斯蒂、龚伯茨和理查德拟合模型.根据各模型的检验参数分析比较4种理论函数对纹枯病进展曲线的拟合性,发现单分子函数的拟合性最差,龚伯茨函数优于逻辑斯蒂函数;当确定适当形状参数m值时,用理查德函数可获得纹枯病发展动态的最优拟合模型,其决定系数最高而卡方值和均方根误差值最小.在实验中设置了3个纹枯病接种处理并获得3条病害进展曲线,用理查德对它们进行拟合所得到的拟合模型分别是:x=[1+0.187 4 exp(-0.213 9 t)]^(-20.5699)±0.0170(m=1.048 6,R=0.982 2,X^2=0.010 9) x=[1-0.690 3 exp(-0.210 3 t)]^(3.6881)±0.0198(m=0.728 9,R=0.988 5,x^2=0.010 9) x=[1+0.636 4 exp(-0.431 6 t)]^(-8.0703)±0.0277(m=1.443 3,R=0.988 2,x^2=0.015 6)A field plot experiment with three inoculum density treatments was conducted to study the development dynamics of rice sheath blight caused by Rhizoctonia solani （anamorph： Thanatephorus cucumeris）. The relative disease severity （x） data of the disease were recorded at different time （t） and crop growth stages during the cropping season and three disease progress curves were produced from these data. The computer software, Epitimulator constructed in our laboratory, was applied for simulation of the disease dynamics. Models simulated with the monomolecular, logistic, Gompertz and Richards functions were obtained by imputing the disease severity data and running Epitimulator. The simulated models were tested for significance and compared for accuracy. Results showed that the simulativity of monomolecular model was the worst and Compertz function was better than logistic function in describing the development curves of sheath blight. However, when the most appropriate shape parameter （m） was found and used, the use of Richards function resulted in the best simulation models of the disease with the highest determination coefficient （R）, the lowest kai-square （x^2） and the smallest root of mean standard error （RMSE）. The simulated models produced for the three inoculation treatments were： x=[1＋0.1874 exp（-0.2139t）]^20.5699±0.0170（m=1.0486,R=0.9822,x^2=0.0109） x=[1-0.6903exp（-0.2103t）^3.6881 ±0.0198（m=0.7289,R=0.9885,x^2=0.0109） x=[1＋0.6364exp（-0.4316t）]^-8.0703±0.0277（m=1.4433,R=0.9882,x^2=0.0156）.

关 键 词：水稻纹枯病 流行时间动态 拟合分析 理查德模型 

分 类 号：S435.111[农业科学—农业昆虫与害虫防治]