The modified Gompertz model developed by Gooding et al. (2000) was used to describe the decline in green area of wheat top three leaves in field experiments where green leaf area at time t = 100*exp[-exp(-k*(t-m))]. In the absence of fungicide, green leaf area decline was associated with drought or infection with a number of foliar pathogens including Septoria tritici (sexual stage Mycospherella graminicola) and Puccinia recondita f.sp. tritici and Erysiphe graminis. On the whole experiments and cultivars there was no effect of fungicide on k but it was highly significant on m (P<0.001). When main effects are considered, fungicide delayed green leaf area decline rather than reduced its rate of progress once started. Fungicide had variable effects on grain yield, largely reflecting variation in disease infection pressures in the different years and susceptibility of the different cultivars. If no correlation was observed between k values and grain yields, m was at the contrary very closely associated with grain yields. This study supports the view of Gooding (2000) and Gaunt (1995) that, for many diseases, effects on green area duration give an adequate estimate of host yield reduction. Therefore we have use the close relationship between the parameter m of the Gompertz model and grain yield to improve the Belgium-Crop Growth Monitoring System (B-CGMS) by recalibrating LAI evolution simulated by B-CGMS according to the evolution of LAI derived from the Gompertz model. The system includes indeed a component allowing taking into account senescence in the evolution of LAI during the growing season. Recalibration has been realized through the modification of one of the parameters influencing the leaf senescence, the SPAN parameter. Modifying the SPAN parameter comes down to modify the lifespan of leaves. Considering that parasitic pressure reduces leaves lifespan and therefore the photosynthetic capacity, this approach makes it possible to take into account the influence of this pressure on yield predictions in ...