Rice blast forecasting models and their practical value: a review

Abstract

Rice, after wheat, is the second largest cereal crop, and is the most consumed major staple food for more people than any other crop. Rice blast (caused by Pyricularia oryzae, teleomorph Magnaporthe grisea) is the most destructive of all rice diseases, causing multi-million dollar losses every year. Chemical control of this disease remains the most effective rice blast management method. Many attempts have been made to develop models to forecast rice blast. A review of literature of the rice blast forecasting models revealed that 52 studies have been published, with the majority capable of predicting only leaf blast. The most frequent input variable has been air temperature, followed by relative humidity and rainfall. Critical factors for the pathogenesis, such as leaf wetness, nitrogen fertilization and variety resistance have had limited integration in the development of these models. This review reveals low rates of model application due to inaccuracies and uncertainties in the predictions. Five models are part of current operational forecasting systems in Japan, Korea and India. Development of in-field rice-specific weather stations, along with integration of leaf wetness and end-user interactive inputs should be considered. This review will be useful for modelers, users and stakeholders, to assist model development and selection of the most suitable models for the effective rice blast forecasting.

Keywords

• leaf disease
• neck disease
• pathosystem
• prediction