Research article

Subplots facilitate assessment of corn yield losses from weed competition in a long-term systems experiment

USDA-ARS Sustainable Agricultural Systems Lab, Building 001 Room 245, 10300 Baltimore Avenue, Beltsville, Maryland 20705, USA

^* Corresponding author: john.teasdale@ars.usda.gov

Accepted: 15 October 2009

Abstract

Weeds can limit crop yield, particularly in organic systems where herbicide technologies are unavailable. Weedy and weed-free subplots were established within the full plots of a long-term Farming Systems Project at Beltsville, Maryland, USA, to determine the effect of weed competition on corn yields in six of the first ten years of an experiment designed to compare conventional and organic cropping systems. Weed abundance was determined in two ways, by sampling above-ground biomass and by estimating percent of soil area covered by weeds. Percent weed cover was shown to be linearly related to weed biomass but was more comprehensive than biomass sampling for determining the overall weed abundance in the larger full plots. The relationship between corn yield and weed cover estimates in subplots was used to estimate corn yield losses to weed competition and weed-free yield in the full plots. Weed competition reduced corn grain yield in all years in subplots, however, the degree of yield reduction ranged from 4 to 76%. This considerable variability was explained by rainfall whereby the highest yield loss occurred in years with below-average rainfall and the lowest yield loss occurred in years with above-average rainfall. Estimation of full-plot corn yield loss to weed competition demonstrated that yield differences between the conventional and organic systems were explained not only by weed competition but also by nitrogen availability. Results suggested that, in years with above-average rainfall, nitrogen availability was more limiting to organic corn yield than weed competition, but that, in years with below-average rainfall, weed competition was more limiting than nitrogen availability.