Forecasting of potato late blight

Late blight of potatoes, the disease responsible for the infamous Irish famine in the mid-nineteenth century, caused by Phytophthora infestans (Mont.) de Bary is still the most dreaded disease throughout the world resulting in losses of over 12 billion annually. These losses can be contained through the use of chemicals (including disease forecasting), cultural practices, and host resistance. Although use of host resistance is the most cost effective management strategy but either resistant varieties are not developed or, wherever available, these cannot withstand the disease pressure on their own. These need to be propped up with some supportive chemical applications. In this context usage of chemicals become inevitable. However, the use of chemicals is costly, there are chances of development of resistance to them, and it is also not environment friendly. To reduce the cost of management and pesticide residues, their need based application is the need of the day. In this context disease forecasting can play an important role.

The principles of disease forecasting are based on relationship between the weather parameters and the disease epidemiology. Van Everdingen (1926) for the first time utilized weather parameters (temperature, RH, rainfall, dew) to develop a late blight prediction system, commonly known as ‘Dutch rules’ for Holland. Since then large number of forecasting systems based on minimum temperature and RH, moving day concept, severity values, risk values etc. have been developed worldwide but most of them are region specific. Bhattacharyya et al. (1982) and Singh et al. (2000) developed forecasting systems for Indian conditions using some of these concepts. Decision support systems like BLITECAST, PhytoPRE, PHYTEB, ProPhy, NEGFRY have also been developed globally for need based application of fungicides. In India JHULSACAST- based decision support system has been developed for different regions of the country. Despite so much development in the field of late blight forecasting, there is still need of a disease forecasting system that can be used globally without any calibration to local situation.