Effect of planter and tractor wheels on row and inter-row weed populations

Abstract Real time weed detection could reduce herbicide inputs in fields but its application is constrained by weed/crop discrimination. This hurdle could be bypassed in crops like corn and soybean if weeds were detected on the inter-row only. However, using inter-row data to control weeds could be risky if weeds are more abundant on rows. A field study was carried out in the province of Quebec (Canada) to evaluate the effect of the planter unit and tractor wheels on relative weed cover, density and biomass between rows and inter-rows in corn and soybean when a post-emergence application of herbicide is scheduled. The experimental design included two sites, two fields per site, two years, and two crops. Treatments included: 1) seeded row (control) 2) unseeded row (seeder pass without crop seeds) and 3) tractor wheel (wheel tracked inter-row). Weed cover was evaluated using digital images and a custom made program. The absence of a crop on rows (unseeded vs. seeded rows) did not modify weed density and biomass before herbicide application in both corn and soybean. Weed density was always higher on rows than on inter-rows unless inter-rows had been tracked by tractor wheels indicating that soil disturbance by the planter and tractor wheels increased weed seed germination and subsequent seedling emergence. Weed cover on undisturbed inter-rows was generally lower or equivalent to weed cover on rows. Weed cover on inter-rows compacted by tractor wheels was always equivalent or higher than weed cover on rows. Using inter-rows to detect weeds could underestimate weed pressure in corn and soybean in Quebec, unless the inter-rows trampled by tractor wheels were scanned.