Weed-induced yield loss through resource competition cannot be sidelined

Abstract

The recent article by Horvath et al. [1.Horvath D.P. et al.Weed-induced crop yield loss: a new paradigm and new challenges.Trends Plant Sci. 2023; 28: 567-582Google Scholar] claims that ‘competition for resources in well-managed agroecosystems is not the primary mechanism underlying the weed-induced crop yield loss’. We do not question the existence of other mechanisms involved (e.g., in perceiving and responding to neighbour plants), and we acknowledge the quality of the paper in describing these mechanisms and in pointing to new research directions. However, the authors’ arguments regarding the secondary role of competition in crop–weed interactions are unconvincing. The term ‘well-managed agroecosystems’ (and synonyms), which is a central concept in the authors’ arguments, was used nine times throughout the article without ever being defined. The authors seem to refer to systems in which resources are sufficient to cover the needs of all plants, at least early in the growing season (Quote 1 in Box 1). However, even in intensive, well-fertilised, and irrigated systems, plant nitrogen deficiencies occur because of nitrogen leaching or gaseous nitrogen losses after excessive rain or because of nitrogen immobilisation in soils [2.Cameron K.C. et al.Nitrogen losses from the soil/plant system: a review.Ann. Appl. Biol. 2013; 162: 145-173Google Scholar]. Incidentally, this focus on nonlimiting resources suggests that organic and conventional low-input systems (among which are integrated weed management systems) or nonirrigated systems are excluded from the scope of this paper. But such systems are widespread all over the world, and their acreage will continue to increase with the ongoing agroecological transition. The authors’ perception of well-managed agroecosystems constitutes a partial vision of agriculture and appears out of line with the ongoing agroecological transitions, including a shift from synthetic inputs (i.e., fertilizers and pesticides) towards organic inputs and biological interactions.Box 1Quotes from the target article that we discuss in the textQuote 1.In well-managed agroecosystems, resources are not generally limited early in the growing season, and water uptake and nitrogen assimilation are highest during the later stages of crop development, peaking during or just prior to grain filling (Figure 3). Yet, surprisingly, weed emergence during these later periods, when resources are most in demand, has minimal impact on yield.Quote 2.If weeds primarily reduced crop yields by limiting resources, then increasing resource inputs should negate yield losses as the resources approach levels that could fully support both crop and weeds.Quote 3.[W]eeds reduced crop yield, often at nearly the same percentage as they did without additional fertilization (Figure 1) [8.Lindquist J.L. et al.Effect of nitrogen addition and weed interference on soil nitrogen and corn nitrogen nutrition.Weed Technol. 2017; 24: 50-58Google Scholar] [and six other references].Quote 4.Even in cases where the nitrogen level in corn was reduced due to weed pressure – for example [8.Lindquist J.L. et al.Effect of nitrogen addition and weed interference on soil nitrogen and corn nitrogen nutrition.Weed Technol. 2017; 24: 50-58Google Scholar], the addition of nitrogen had little impact on the percentage of nitrogen lost due to weed presence, suggesting that weeds were impacting nitrogen assimilation in the crop, regardless of the level of available nitrogen.Quote 5.Weeds that emerge early relative to the crop have the greatest impact on yield, whereas weeds that emerge during the later stages of crop vegetative growth have minimal impact on crop yield, even if they overtop the crop.Quote 6.If weeds were reducing yield by direct competition for resources, then one would expect a more linear impact on yield as weed biomass increased. Instead, […] the rate of yield loss per unit weed is higher with low weed levels and decreases as weed density or biomass increases. Quote 1.In well-managed agroecosystems, resources are not generally limited early in the growing season, and water uptake and nitrogen assimilation are highest during the later stages of crop development, peaking during or just prior to grain filling (Figure 3). Yet, surprisingly, weed emergence during these later periods, when resources are most in demand, has minimal impact on yield.Quote 2.If weeds primarily reduced crop yields by limiting resources, then increasing resource inputs should negate yield losses as the resources approach levels that could fully support both crop and weeds.Quote 3.[W]eeds reduced crop yield, often at nearly the same percentage as they did without additional fertilization (Figure 1) [8.Lindquist J.L. et al.Effect of nitrogen addition and weed interference on soil nitrogen and corn nitrogen nutrition.Weed Technol. 2017; 24: 50-58Google Scholar] [and six other references].Quote 4.Even in cases where the nitrogen level in corn was reduced due to weed pressure – for example [8.Lindquist J.L. et al.Effect of nitrogen addition and weed interference on soil nitrogen and corn nitrogen nutrition.Weed Technol. 2017; 24: 50-58Google Scholar], the addition of nitrogen had little impact on the percentage of nitrogen lost due to weed presence, suggesting that weeds were impacting nitrogen assimilation in the crop, regardless of the level of available nitrogen.Quote 5.Weeds that emerge early relative to the crop have the greatest impact on yield, whereas weeds that emerge during the later stages of crop vegetative growth have minimal impact on crop yield, even if they overtop the crop.Quote 6.If weeds were reducing yield by direct competition for resources, then one would expect a more linear impact on yield as weed biomass increased. Instead, […] the rate of yield loss per unit weed is higher with low weed levels and decreases as weed density or biomass increases. Much more important, the authors’ vision of crop–weed interactions tends to overlook a major resource for which crops and weeds compete in temperate and tropical arable cropping systems (i.e., light) [3.Cahill J.F.J. Interactions between root and shoot competition vary among species.Oikos. 2002; 99: 101-112Google Scholar, 4.van Breugel M. et al.The relative importance of above- versus belowground competition for tree growth during early succession of a tropical moist forest.Plant Ecol. 2012; 213: 25-34Google Scholar, 5.Wilson S.D. Tilman D. Plant competition and resource availability in response to disturbance and fertilization.Ecology. 1993; 74: 599-611Google Scholar]. The authors almost exclusively focus on soil resources; light is only considered in terms of quality (e.g., red to far red ratio), involved in the perception of neighbouring plants. The paper ignores that the growth of individual plants in communities is always constrained by limiting factors, and, when soil resources are plentiful, light becomes the limiting resource for which plants compete. Disregarding this basic concept from ecology results in biased conclusions. For instance, the authors conclude several times that there was no crop–weed competition for resources because nonlimiting or additional soil resources did not reduce weed-caused yield losses (e.g., Quote 2–Quote 4). However, this does not preclude plant–plant competition for light, and increasing nitrogen availability can actually increase this competition, notably for understory plants of canopies eutrophicated by nutrient enrichment [6.Eskelinen A. et al.Light competition drives herbivore and nutrient effects on plant diversity.Nature. 2022; 611: 301-305Google Scholar,7.Hautier Y. et al.Competition for light causes plant biodiversity loss after eutrophication.Science. 2009; 324: 636-638Google Scholar]. Moreover, several of the studies cited by the authors to support their reasoning (e.g., in Quote 3–Quote 4) actually concluded that competition for resources other than nitrogen was the dominant mechanism (e.g., [8.Lindquist J.L. et al.Effect of nitrogen addition and weed interference on soil nitrogen and corn nitrogen nutrition.Weed Technol. 2017; 24: 50-58Google Scholar] concluded that 60%–71% of the variance in corn shoot biomass must be explained by factors such as competition for resources). Horvath et al.’s extensive investigation of fine-scale plant–plant interactions other than competition tends to miss part of the complexity of interactions in cropped fields. For instance, the authors argue against a key role of competition for resources by stating the well-reported fact that earlier emerging weeds are more harmful for crop yield than late-emerging weeds (e.g., Quote 5, Quote 1). But, similarly to the understory plants in the eutrophicated canopy of [6.Eskelinen A. et al.Light competition drives herbivore and nutrient effects on plant diversity.Nature. 2022; 611: 301-305Google Scholar,7.Hautier Y. et al.Competition for light causes plant biodiversity loss after eutrophication.Science. 2009; 324: 636-638Google Scholar], the latecomers lack the light quantity to grow and thus require few soil nutrients. In other words, competition is dictated much earlier than when it takes place. Moreover, competition occurs earlier than the figure cited in Quote 1 suggests. This figure shows nitrogen accumulation in maize, which indeed peaks during reproduction. But the rate of nitrogen uptake (i.e., the slope of accumulation vs. time), and thus the time when plants actually compete for the resource, is highest during vegetative growth. Finally, the authors conclude that the absence of a linear relationship between weed biomass and weed-induced yield loss supports the predominance of mechanisms other than resource competition (e.g., Quote 6). But linear models are not suitable across a wide range of weed densities, because they lack the concavity to represent interspecific and intraspecific interactions and may overestimate yield loss [9.Ali A. et al.Yield loss prediction models based on early estimation of weed pressure.Crop Protect. 2013; 53: 125-131Google Scholar]. Even simulations with models considering only crop–weed competition for light show that the correlation between yield loss and weed biomass is S-shaped (i.e., sigmoid) and not linear [10.Colbach N. Cordeau S. Reduced herbicide use does not increase crop yield loss if it is compensated by alternative preventive and curative measures.Eur. J. Agron. 2018; 94: 67-78Google Scholar]. Indeed, because light (in contrast to soil resources) is mostly monodirectional, competition for light is not symmetric; that is, tall individuals capture proportionally more light than smaller ones [11.Schwinning S. Weiner J. Mechanisms determining the degree of size asymmetry in competition among plants.Oecologia. 1998; 113: 447-455Google Scholar,12.Bourgeois B. et al.What makes a weed a weed? A large-scale evaluation of arable weeds through a functional lens.Am. J. Bot. 2019; 106: 90-100Google Scholar]. In conclusion, in our opinion, considering competition for light and soil resources in crop–weed interactions is essential for agroecological weed management, and disregarding this process weakens some of the key messages supported by the authors, such as the need to breed for weed-tolerant crop varieties. Although we agree that mechanisms other than resource competition exist and should be explored, they cannot be investigated by sweeping other key mechanisms under the carpet.

Topics & Concepts