Predictive analysis of cotton leaf curl disease using growth models in Haryana's Hisar Region

Cotton leaf curl virus disease (CLCuV) incidence has increased in the northern states of India (Punjab, Haryana, and Rajasthan) since 1993.CLCuVreducesthenumberofharvestablebollsby15-87 % and seed cotton yield by 11-92%, depending on the extent and timing of Infection.Cotton leaf curl virus disease (CLCuV) is among the most damaging cotton diseases.It was first found on Gossypium barbadense in Nigeria (Africa) in 1912. CLCuV was first reported in India in Sri Ganganagar (Rajasthan) in 1993 and in Punjab (India) in 1994. Various mathematical models have been employed to create models which describe epidemic dynamics.In this study, an attempt was made to compare three models —namely, the Monomolecular, Logistic, and Gompertz models —using secondary data from 2017 to 2022 for the prediction of Cotton Leaf Curl Disease (CLCuD).The studied models were evaluated using goodness-of-fit criteria, i.e., the Coefficient of determination and root mean square error.It was observed that the Coefficient of Determination (R²) was found to be in the range of 0.98 to 0.99 for all years,whereas the root mean square value was observed in the range of 2.06 to 16.25.The value of the Coefficient of determination (R²) was very high, and the mean square error was very low for the years 2017, 2019, 2020, and 2022, respectively, for the logistic model compared to other studied models. It was observed that 98 to 99 % the variation in disease intensity was explained by time, and an error was also observed to be minimal for all years.It was also concluded that, for the years 2017 to 2022, except for 2018 and 2021, the logistic model was found to be the best fit for predicting disease severity. In contrast, for 2018 and 2021, the Gompertz model was found to be the best. The analysis was performed using R and Excel software.