Aim: This research was carried out to identify the weather-based prediction model for rice yellow stem borer (YSB) population. The Stemborer incidence was recorded by using the light trap installed at TamilNadu Rice Research Institute (TRRI). The YSB data and climatological data of Aduthurai centre (Cauvery delta zone) from 2012 to 2022 were considered for this work. Methodology: The research methodology involved the conversion of daily light trap catch data into weekly data, which was then organized according to the Standard Meteorological Week (SMW) for model development. Three different predictive models were developed: Multiple Linear Regression (MLR), Stepwise Multiple Linear Regression (SMLR), and ARIMAX models. These models were used to forecast outcomes based on the weekly weather data and light trap catches. Results: The results indicated that the ARIMAX model demonstrated the lowest Root Mean Square Error (RMSE) when compared to both MLR and SMLR models, showing superior performance overall. Higher maximum and minimum temperatures negatively affected the YSB population, with decrease of 0.34 and 2.10 units per degree increase, respectively. Conversely, relative humidity had a positive effect, with an increase of 0.29 units per degree rise. This model underscores the significant influence of temperature and relative humidity on the population dynamics of stemborer. Interpretation: The model developed by using ARIMAX performed well than other models. It helps to conclude that weather parameters have can influence on pest population. Moreover, the models created in this research will significantly aid for monitoring and management purpose of the Yellow stem borer population in rice crop. Key words: ARIMAX, Oryza sativa, Rice yellow stem borer, Scirpophaga incertulas, Weather variables
Read full abstract