A computer program using the language and statistical procedures available from SAS (Statistical Analysis System) was written in order to identify the most highly correlated meteorological factors with the incidence of wheat head blight (caused byFusarium graminearum Schwabe) at Pergamino, in the humid pampeana region. Applying linear regression techniques, different models from simple up to a maximum of three independent variables were fitted to the data (1978–1990). The meteorological variables were processed in a time segment beginning eight days prior to the heading date (50% of emerged ears) and finishing when 530 degree days were accumulated (26–32 days). The number of two day periods with rainfall and relative humidity >81% in the first day and relative humidity ≥78% in the second (NPPRH) was the variable that showed the strongest association with disease incidence (FI) (R2=0.81). After examining the models in several ways (R2, Adjusted R2, PRESS statistic), two equations were selected: FI%=20.37 + 8.63 NPPRH − 0.49 DDXNT (R2=0.86) and FI%=16.39 + 5.43 NPPRH − 0.45 DDXNT + 2.95 DPRH (R2=0.886), in which DDXNT represents the daily accumulation of the residuals resulting from subtracting 9 to the minimum temperature values (<9 °C) and the exceeding amounts of maximum temperatures from 26 ° C and DPRH is the number of days with precipitation and relative humidity greater than 83%. Successful local predictions of incidence of scab for the years 1991–1993 (reserved for validation purposes) were achieved using both equations.
disease incidence Fusarium graminearumhead blight or scab linear regression models meteorological variables