Forecasting Septoria nodorum on winter wheat in England and Wales

Abstract

SUMMARY: The regional incidence of septoria (Septoria nodorum Berk), on winter wheat, as found in the National Winter Wheat Leaf Disease Survey for 1970–75, was related to weather variables and amounts of inoculum. The relation between disease shortly before harvest and days with rain through the growing season was explored by a correlation method and it was found that the closest association was with days with 1 mm or more of rain (wet days) in the period mid‐May to mid‐June, When explored in detail, the relation between infection and wet days showed both non‐linearity and asymmetrical scatter, A graphical technique was therefore used to develop simple forecast rules. These performed well when tested against data for 1976–78, although these years did not include a septoria epidemic.In addition to wet days, the influence of other weather variables was investigated by a multiple regression technique. Sunshine in May and June and temperatures in July were found to have a slight effect on septoria incidence, as did the previous year's infection (taken as a measure of inoculum) but the overall reduction of variance was slight. It is suggested that these additional variables, together with others not treatecl quantitatively, may best be used to decide the regional forecast in marginal cases, and to make it applicable to individual crops.Attempts to relate septoria at individual sites in northern England to wet days at nearby rainfall stations were not successful; nor did rainfall data from all these stations contribute to a better regional forecast of septoria than did a few stations chosen at random.