Simulation of above-ground suppression of competing species and competition tolerance in winter wheat varieties

Cereal crop species and varieties differ in competitive ability against weeds mainly as influenced by differences in canopy architecture. The FASSET crop model was used to separate the effects of a number of crop traits on the suppressive ability of winter wheat varieties and the ability to tolerate weeds. The model simulated the competition between different varieties of winter wheat and a sown grass mixture for light, water and nitrogen. Crop physiological parameters of eight varieties and one variety mixture were estimated from measurements in a 3-year field experiment. The parameters estimated were thermal time from emergence to flag leaf appearance, thermal time from flag leaf appearance to anthesis, thermal time from anthesis to yellow ripeness, height development rate and final plant height, specific leaf area, leaf area per N uptake, vertical displacement of leaf area, and extinction coefficient for light. Variety differences were found in most of the parameters. The varieties were evaluated in terms of their abilities for suppression and tolerance. A suppression index (SI) was defined as the ratio of grass biomass at start of grain filling in a specific variety divided by the average biomass over all cultivars. A tolerance index (TI) was defined as the slope of a linear regression between crop yield and weed biomass at anthesis. Significant variety differences were observed for SI, but not for TI. The simulation model was able to capture about 75% of the variation in observed SI and about 47% of the variation in TI. The suppression and tolerance indices were highly correlated both for the observed and simulated data. A model sensitivity analysis documented that several crop traits were affecting both SI and TI leading to this correlation. The crop traits that affected the suppression index most in the simulations were early crop development, rapid height growth and specific leaf area. The response of the tolerance index to variation in crop characters was smaller and also related to factors such as the extinction coefficient and leaf area per unit N uptake. The suppressive ability of the winter wheat varieties were found in the simulations to have a smaller inter-annual variability than the ability to tolerate a given weed pressure. The results show that a simulation model may be used in selecting for competitive genotypes by measuring a few key variety traits. This may be done either by direct use of the model or indirectly through the development of simpler indices based on the simulation model. It is further proposed that a simulation model may be an effective tool in including crop competitiveness in the official variety approval since the measurements of the variety dependent input parameters defining competitiveness may be less costly than performing variety trials under a range of weed densities. (C) 2004 Elsevier B.V. All rights reserved.