Investigations into Alternative Methods to Predict the Competitive Effects of Weeds on Crop Yields

Abstract

Sixteen experiments have investigated alternative methods of predicting the competitive effects of a simulated weed (oats) on the yields of spring barley, spring oilseed rape (canola), peas, spring field (faba) beans and flax. The experiments were designed to discover whether early postemergence assessments of crop and weed vigor would achieve more reliable prediction of yield loss than weed density. Weed density (plants m−2) was a very variable predictor of yield loss. The standardized ranges (range/mean) of values over 3 to 4 years of data for the five crops, in the densities causing 5% yield loss, were between 1.14 and 2.59. Predictions based on the relative dry weight of crop and oats (oat dwt/(oat dwt + crop dwt)), assessed while the plants were still small, achieved more reliable predictions, as the standardized ranges were between 0.10 and 1.86. In three of the experiments, predictions based on relative dry weights were compared to similarly timed predictions based on measurements of relative leaf area and of ground cover, assessed subjectively (by eye) and photographically. Subjective and objective (photographic) assessments of cover achieved similar predictions of yield loss, indicating that visual assessments could be a viable tool to assess the potential competitive effects of weeds.

Prediction of competition between oilseed rape and Stellaria media

“…The predictive techniques of Kropff & Spitters (1991) were devised in sugarbeet and rice. Our confirmatory work was done in field beans ( Lutman et al ., 1996 ). All these are short‐season crops, and the opportunity for post‐prediction climatic variation to influence the predicted yield responses is low.…”

“…(1995 ) addressed this issue by predicting yield loss from assessments of relative weed leaf area made early in the life of the crop. We have used these models in spring crops such as field bean ( Vicia faba L.) and have confirmed that more robust predictions are possible using the Kropff models than when using the weed density approaches ( Lutman et al ., 1996 ; Van Acker et al ., 1997 ). However, the conclusions from the work in this paper on S. media in oilseed rape are less positive.…”

Diagnosis of limiting factors of organic oilseed rape yield. A survey of farmers’ fields

“…We determined total aerial dry biomass of all weeds and total nitrogen content of the aerial parts of the plant for each sample (Dumas method). Two indicators of weed competition were calculated: (1) the relative weed biomass ratio = DM weeds/(DM weeds + DM crop), where DM is Dry Matter as used by Lutman et al (1996); and (2) the relative weed nitrogen ratio = QN weeds/(QN weeds + QN crop), where QN is Quantity of Nitrogen in the plants. Other indicators included weed species, density of each species, total density of weeds and the type of weed with the highest density.…”

“…These different indicators of weed competition (the relative ratio of weed dry biomass, the relative ratio of weed nitrogen and weed density, and weed density by type) must be discussed a little. The first of these indicators has been used to assess weed competition in previous studies (Lutman et al, 1996(Lutman et al, , 2000. Lutman et al (2000) showed that yield loss predictions based on relative weed dry weight were more accurate than those based on weed density.…”

Competividade e predição de perdas de rendimento da cultura de feijão quando em convivência com Brachiaria plantaginea (Link) Hitchc.

“…O ajuste dos dados obtido com modelo que utiliza S pd como variável [eq. (30] (Lutman et al, 1996). Os valores de perda de rendimento da cultura (Vicia faba) estimados com os valores das avaliações visuais de S pd foram semelhantes entre os quatro avaliadores (r 2 variou entre 0,78 e 0,83) e semelhante aos obtidos com a avaliação por meio de fotos (r 2 = 0,87).…”