Evaluation of a mechanistic model of wheat and weed growth

Abstract

A mechanistic crop growth model, where daily growth is the product of light use efficiency and intercepted radiation, was tested for its ability to simulate biomass growth in wheat and weeds. Wheat was sown at six densities (including 0), in September 1994 in Canterbury and the naturally occurring weed community was either left intact or removed by herbicide. Weed biomass growth was accurately simulated in the presence and absence of the crop, but late-season wheat growth was underestimated, particularly at low sowing densities. Herbicide treatment reduced early-season crop growth, but a grain yield loss of 6.8% due to weed competition, was prevented. Weed biomass accumulation was reduced with increasing wheat density.