Determination of the Best Forecast System for the Prediction of the Tomato Late Blight in the Field Tomato Growings of Balıkesir and Çanakkale provinces

A large number of late blight forecasting models exist and are currently used in many tomato and potato production areas throughout the World. The use of forecast systems to predict disease has the potential of reducing fungicide applications without reducing yield. This study carried out in the tomato growing areas of Balikesir (Merkez, Manyas, Susurluk) and Canakkale (Batak ovasi) provinces between the years of 2006-2010. In the region where the automatic weather stations were located no fungicide applications were carried out in the control tomato fields until the occurrence of the first symptoms of the disease on the plants. Surveys were carried out once and sometimes twice a week  to determine  the occurence of the first symptoms of the disease. Comparisons were made between the predictions  of the  model and the actual date of the occurence of the  blight late in the field ,with aim to verify the suitibility of the tested  model. In study had used SMITH, Fry Phytophtora Unit, IPI, TOMCAST and Tomcast model was used with modification as named modified TOMCAST forecasting models which are predict infection conditions of tomato late blight, caused by Phytophthora infestans. These models were compared between the actual date when blight was found in the crop and predict infection date of these models. Thus, The most appropriate models were determined according to study areas. After result of study, modified TOMCAST model were found to be the suitable at Balikesir regions. and SMITH model were found to be the suitable at Canakkale regions.