PestPredict-EMS (Kharif)

病蟲害預測經驗模型的系統在秋收季節

Forewarning is an essential component of IPM, and predicting of pest incidence based on changing climate is a challenge as its impact on pests is both direct and indirect. e-pest surveillance done under CROPSAP and NICRA has sustained and widened the scope of forewarning. Given the broad range of forecasting methods, rule-based predictions have served as simple but robust tools of forewarning of rice insect pest severity amongst seven locations viz., Ludhiana (Punjab), Chinsurah (West Bengal), Raipur (Chhattisgarh), Karjat (Maharashtra), Hyderabad (Telangana), Mandya (Karnataka) and Aduthurai (Tamil Nadu) besides maximum severity of Spodoptera litura for groundnut cropping system at Dharwad (Karnataka) and early blight of tomato at Bengaluru (Karnataka). Empirical models using field incidence/severity of insect pests and diseases of rice, pigeonpea, groundnut and tomato for various locations with their respective weather patterns have been validated to be effective for designated periods of crop season. Such empirical models and the rule based models have been facilitated for validation as well as field use through web hosting as well as application for use through web and smart phones.

PESTPREDICT, an android mobile application for pest forewarning assists researchers, extension personnel of agriculture and farmers to get location specific forecasts of desired insect pest(s) or disease(s) for their effective management on target crops. PESTPREDICT reduces calculation efforts and provides an instant and extempore framework for use of developed prediction models. While success of predictions can be continuously improved through refinements considering the changing pest scenario and climate, immediate requirement is the need for locations of target crops to use the application on regular basis and issue ‘pest alerts’ at times prediction of ‘high’ severity levels. Web enabled/mobile application based predictions are specific to locations and insects, creating awareness amongst potential users including extension functionaries is of utmost importance.

It is always pertinent to have a ‘pest alert’ backed up by direct and random field level monitoring.