Relación entre la precipitación registrada en estados reproductivos de la soja y la severidad de Septoria glycines y Cercospora kikuchii

• Autores: Carmona, Marcelo Aníbal; Moschini, Ricardo; Cazenave, Graciela; Sautua, Francisco José
• Año de publicación: 2010
• Idioma: español castellano
• Tipo de recurso: artículo
• Fuente: Tropical Plant Pathology; Vol.35, no.2; 71-78; http://www.sbfito.com.br/
• Materia: GLYCINE MAX; LATE SEASON DISEASES; LOGISTIC MODELS; CERCOSPORA; CERCOSPORA KIKUCHII; PUNCTUM BLANDIANUM; SEPTORIA GLYCINES
• Institución: Universidad de Buenos Aires. Facultad de Agronomía

Fil: Carmona, Marcelo Aníbal. Universidad de Buenos Aires. Facultad de Agronomía. Departamento de Producción Vegetal. Cátedra de Fitopatología. Buenos Aires, Argentina.
Fil: Moschini, Ricardo. Instituto Nacional de Tecnología Agropecuaria (INTA). Instituto de Clima y Agua. Castelar, Buenos Aires, Argentina.
Fil: Cazenave, Graciela. Instituto Nacional de Tecnología Agropecuaria (INTA). Instituto de Clima y Agua. Castelar, Buenos Aires, Argentina.
Fil: Sautua, Francisco José. Universidad de Buenos Aires. Facultad de Agronomía. Departamento de Producción Vegetal. Cátedra de Fitopatología. Buenos Aires, Argentina.
From 2003 to 2008, late season disease severity (LSDsev) values (Brown spot and Cercospora leaf blight) were recorded at the R7 growth stage at several sites in Santa Fe and Córdoba Provinces. The annual LSDsev records (N=15) were grouped into two epidemic categories based on a threshold value (median of observed disease data) : severe (LSDsev less than 36 percent) and moderate to light (LSDsev more or equal to 36 percent). Variations in epidemic levels were studied in relation to precipitation-based variables, processed in a time window limited by R3 and R5 growth stages. The variables were expressed as frequency (days) and total accumulation (mm) of daily precipitations greater than the thresholds 1, 5, 7 and 10 mm. The variables PrF7 (number of days with precipitations less than 7 mm) and AcPr7 (total accumulation of daily precipitations less than 7 mm) and the interaction (product) between them (It7) were the most strongly correlated, according to Kendall tau-b coefficients (rK=0.74, 0.60 and 0.71 respectively).Logistic regression model including AcPr7 correctly estimated the probability of occurrence of epidemic categories in 12 cases (out of 15). Logistic models integrating PrF7 or the interaction effect (It7) presented prediction accuracies of 93.3 percent. These results could be useful for prediction and chemical control of LSD.