Predictive modeling for chickpea blight (Ascochyta rabiei) occurrence in the semi-arid zone using meteorological data from Faisalabad, Pakistan

Chickpea blight is the most destructive disease in the semi-arid zone of Punjab and is mainly controlled through fungicides. However, in this area, the use of fungicides is excessive and non-judicious which could be rationalized through the use of a predictive model based on meteorological variables. The aim of the current research was to develop a disease predictive model of chickpea blight based on temperatures (maximum and minimum), rainfall, relative humidity (RH), and wind speed. Relationship of meteorological variables with disease severity was determined through correlation analysis, and stepwise regression was used to develop the model. For this purpose, 2 yr (2011-12) data of meteorological variables and chickpea blight severity was used. A significant correlation was found between all environmental variables and blight severity. A model based on weekly meteorological variables fit the data well (R2 = 0.82). Predictions of the model were evaluated on two statistical indices, root mean square error (RMSE) and error (%), which were less than or equal ± 20, indicating that the model was good. The model was validated with 5 yr (2006-10) independent data set. Homogeneity of the regression equations of the two models, 2 yr (2011-12) and 5 yr (2006-10), showed that they validated each other. Scatter plots showed that blight severity was high at maximum (20-24 deg C) and minimum (12-14 deg C) temperatures, 65-70% RH, 5-6 mm rainfall and 5-6.5 km/h wind speed). The chickpea blight model developed during this study is the first meteorological variable model in the semi-arid zone of Punjab and will help to make the predictions of chickpea blight well before the occurrence of the disease; thus, the model can make early an prediction of the time of fungicide application, lessen the use of fungicides, curtail input cost of farmers, and help to mitigate environmental pollution.