Hugo Salinas

Background: Spatially targeted spraying is an alternative to broadcast herbicide application that reduces herbicide use by restricting treatment to areas where weeds are detected. However, weed detection using remote or proximal sensing and classification algorithms is imperfect, which may allow weeds to escape treatment. This study examined the evolutionary consequences of spatially targeted spraying and evaluated whether applying herbicide to spray‐buffer zones surrounding detected weeds could compensate for reduced detection accuracy. We tested two hypotheses: first, that spray‐buffer zones can offset imperfect weed detection; and second, that lower detection accuracy and smaller spray‐buffer zones accelerate crop failure, by increasing weed survival, resistance evolution and population growth. Results: A new spatially explicit simulation model was used to represent weed populations with defined genetic architectures conferring different levels of herbicide resistance. The simulation results supported the first hypothesis, showing that spray‐buffer zones reduced the proportion of weeds escaping treatment under imperfect detection. However, the second hypothesis was not supported. More‐accurate spraying removed susceptible weeds more effectively but also imposed stronger selection for herbicide resistance. By contrast, less‐accurate detection resulted in larger weed populations, maintained a higher proportion of susceptible genotypes, and slowed the evolution of resistance. Conclusion: Spray‐buffer zones can partly compensate for imperfect weed detection in spatially targeted spraying systems. However, improving spray precision does not necessarily delay future crop failure from weed competition, because stronger short‐term control may accelerate resistance evolution. These findings highlight an important trade‐off between immediate herbicide efficacy and the long‐term durability of weed management strategies. They also underscore the importance of diversifying weed management strategies alongside the development of more‐accurate targeted herbicide application technologies. © 2026 The Author(s). Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Aims Although several studies have shown that species interactions can support coexistence of plant species, our understanding of the role of root traits and architecture in coexistence is limited, despite their key role in resource acquisition. Theory suggests two seemingly contrasting outcomes: (1) when species share similar traits, competitive exclusion is slowed and persistence is prolonged, or (2) when species differ greatly in their traits, niche separation reduces competition allowing longer persistence. Here, we ask: Does greater divergence in root architecture increase or decrease persistence time in simulated plant communities? Methods We use a functional-structural plant model to simulate the development of annual plant communities over time. The plants belong to one of five species that differ in root architecture, resulting from simulations of evolution under varying levels of competition. Results Our findings support both theoretical predictions, but through different mechanisms. Differences in root morphology increase persistence time, with the two most different species persisting indefinitely. This appears to result from a density-dependent effect that enables species to recover from low densities. In contrast, in communities composed of ecologically identical species, persistence time seems to be governed by neutral drift and increased with their population density and area size. Conclusions The magnitude of differences in root morphology of interacting species drive their expected persistence time through distinct mechanisms.

Plants' competitive ability involves both suppressing the growth of neighbours (competitive effect) and resisting or tolerating their suppression (competitive response). Competition for below‐ground resources must be related to the ability of plants to acquire these resources, which is mediated by roots and their morphology. However, the role of root architecture in the competitive ability of plants, and in the possible trade‐offs among growth potential, competitive suppression and competition tolerance involved, has not been extensively studied. We used a functional‐structural root model coupled with an evolutionary algorithm to simulate the evolution of root architectures in five scenarios with different plant densities. We asked (1) does selection under different intraspecific competition scenarios result in different root architectures? and (2) do differences in these architectures result in differences in growth potential and competitive ability, that is, competitive effect and response? Our results indicate that as the number of neighbours increases, selection on traits such as branching angles, gravitropism and branching probability results in root architectures that are deeper and sparser, resulting in lower shoot biomass. We also found a difference in competitive ability among architectures, with a trade‐off between resistance to competition on one hand, and competitive effect and maximum productivity (maximum shoot biomass) on the other: there is not a globally optimal strategy. Our findings have implications for management of invasive species, improvement of crop yield and the study of species co‐existence.

We aimed to (i) develop a new computational functional–structural plant root model (FSPM) able to represent a wide range of realistic root system structures and growth strategies with varying levels of phenotypic plasticity while being flexible and efficient enough to facilitate our longer term objective of simulating plant populations or communities interacting and evolving over many generations in different conditions and (ii) use this model to study the evolution of root architecture and plasticity in environments with different levels of soil heterogeneity and rainfall frequency. A model was designed to meet the requirements above. We successfully tested the new model by using it together with an evolutionary algorithm to simulate the evolution of populations of plants with asexual reproduction growing independently. Our model was able to represent realistic root system structures and growth strategies while being flexible and efficient enough to simulate plant populations or communities interacting and evolving over many generations in different conditions. Over evolutionary generations, structural root growth strategies developed and overall fitness (measured as shoot biomass) improved. Different strategies evolved in different evolutionary runs and in different conditions. Evolution clearly selected for root branching plasticity in response to increased local soil water content, indicating that this plasticity confers fitness benefits due to increased water uptake. We conclude that dynamic evolutionary FSPMs have the potential to help us understand the eco-evolutionary costs and benefits of different functional–structural growth strategies. Spatial and temporal heterogeneity in water availability causes the evolution of contrasting root architectures, with different levels of plasticity.