Most predictions of crop yield loss due to weeds are based on a regular or uniform spacing of weeds. When weed distribution is patchy, as in most crop fields, overestimation of yield loss may occur and predicted economic weed threshold densities may be too low. We examined the influence of regular, random and clumped distributions of barnyardgrass onits interference with direct-seeded tomatoes. No comparable studies with these or other species have been reported in the literature to date.|
At a commercial planting density, tomato yield losses in 1994 ranged from 8 to 50% for the clumped arrangement and from 11% to 75% for the regular and random arrangements, respectively, for equivalent barnyardgrass densities. Yield loss differences were related to changes in the proportion of weed-weed and weed-crop competition associated with the different spatial distributions of barnyardgrass. Competition of barnyardgrass with itself increased 14 and 49%, respectively, as the spatial arrangement was changed from regular to random or from regular to clumped. In contrast, the intensity of competition experienced by tomato from barnyardgrass decreased by 8 and 78%, respectively, for the same changes in barnyardgrass aggregation. Changes in the amount of shading of the tomato canopy by barnyardgrass contributed to yield loss differences for the various spatial arrangements. Clumped barnyardgrass caused significantly less shading than barnyardgrass in regular or random arrangements.
Compared to the regular and random arrangements, barnyardgrass growth was substantially reduced in the clumped spatial arrangement, primarily due to increased competition of barnyardgrass with itself. Competition from tomato suppressed seed production of barnyardgrass in 1993. Averaged over all spatial arrangements, seed production was reduced about 60, 70 and 80% by tomato densities of 5, 10 and 20 plants/m, respectively. With less competition from tomato in 1994, suppression of barnyardgrass seed production is expected be similarly reduced.
According to yield loss/weed density relationships developed in this study, predicted economic threshold densities of barnyardgrass for a commercial planting density of tomato (10 plants/m) would then be 104, 130, and 186 plants/acre, respectively, for regular, random and clumped spatial distributions of barnyardgrass. Based on these results, a producer might accept a somewhat larger average density of barnyardgrass before deciding to implement weed control if the barnyardgrass were growing in patches as opposed to being more evenly distributed over the field. However, seed production from even the lowest density of barnyardgrass, which exceeded 100 million per acre and at which spatial arrangement was not critical, necessitates that long-term thresholds for this weed be less than one plant per acre.