Development and evaluation of a model for predicting <i>Lolium rigidum</i> emergence in winter cereal crops in the Mediterranean area

Summary Lolium rigidum is an extremely competitive and prevalent grass weed in cereal fields of Mediterranean areas. The proper timing of control measures is a prerequisite to maximising herbicide efficacy, in terms of both improved control and reduced herbicide inputs. The development of models to predict emergence flushes will contribute to this goal. Pooled cumulative emergence data obtained during three seasons from a cereal field were used to develop a Gompertz model. This explained relative seedling emergence from crop sowing onwards as a function of: (i) standard soil thermal time accumulation ( TT ) with a base temperature of 1.8°C and (ii) soil thermal time accumulation corrected for soil moisture ( cTT ). For the latter, no thermal time accumulation was computed for days in which the soil water balance within the upper 10‐cm soil layer indicated no water available for plants, because evapotranspiration was greater than rainfall plus the stored water remaining from the previous day. The model was validated with six datasets from four different sites and seasons. Compared with TT , the model based on cTT showed better performance in predicting L. rigidum emergence, particularly in predicting the end of emergence. Complemented with in‐field observations to minimise deviations, the model may be used as a predictive tool to better control this weed in dryland cereal fields of Mediterranean climate areas.