Controle da requeima em batata cv. 'Asterix' como base para modelos de previsão da doença.

The control potato late blight needs a great number of fungicide sprayings. These increase the costs of cropping and cause undesirable environmental impacts. The use of forecast systems to predict disease has the potential of reducing fungicide applications without reducing yield. The objective of this study was to evaluate the performance of Blitecast and Prophy systems as a reference model for predicting late blight potato and support decision of spray fungicides. Experiments were carried out during Spring 2004 and Autumn 2005, in Santa Maria, RS, Brazil. Meteorological data were measured in the center of the experimental area at 0.10 and 1.50 m above soil surface. Different accumulated severity values (VS) were calculated with 'Blitecast' (VS = 18, 24, 30, 36 and 42) and 'Prophy' (VS = 15, 20, 25, 30 and 35) forecast systems. These values were used to perform. Two additional treatments were the weekly sprays and without fungicides. The experimental design was a completely randomized, with four replications. Each plot had four rows plants with 5 m length. Late blight severity was evaluated in each three to five days. The fungicide spraying schedule based on Blitecast system with 18 accumulated severity values increased tuber yield at least 42.6 % compared to the control without fungicides treatment. It also reduced the number of sprayings by 25% during wet periods and 70% during dry periods compared to weekly sprayings. The efficiency of controlling late blight was similar to the weekly sprayings treatment in the treatments Bli18 e Pro15. The Prophy model predicted higher number of fungicide sprayings than the Blitecast system and did not improve disease control. [ABSTRACT FROM AUTHOR]