Nonlinear statistical models for plant disease.
Chandran K. P., Prajneshu
Author Affiliation: Indian Agricultural Statistics Research Institute, New Delhi 110 012, India.
Indian Journal of Agricultural Sciences 75 : 361-363
Abstract : The results are presented of a study on the development of powdery mildew (Oidium mangiferae) on mango (cultivars Dashehari and Amrapali) using generalized nonlinear growth models, i.e. monomolecular, logistic and Gompertz models. Convergence of the parameter estimates did not take place when the generalized monomolecular and Gompertz models were used. The generalized logistic model provided a good fit to the given datasets. Further, the AUDPC values for powdery mildew disease with respect to the data of Dashehari and Amrapali were computed as 63.96 and 41.12, respectively, indicating that the former is less resistant to powdery mildew than the latter. The graphs for the rate of disease growth for the generalized logistic model showed that fungicides may be applied at 6 and 3 weeks after the onset of disease on Dashehari and Amrapali, respectively.