Theoretical assessment of persistence and adaptation in weeds with complex life cycles

Abstract

Herbicide-resistant weeds pose a substantial threat to global food security. Perennial weed species are particularly troublesome. Such perennials as Sorghum halepense spread quickly and are difficult to manage due to their ability to reproduce sexually via seeds and asexually through rhizomes. Our theoretical study of S. halepense incorporates this complex life cycle with control measures of herbicide application and tillage. Rooted in the biology and experimental data of S. halepense, our population-based model predicts population dynamics and target-site resistance evolution in this perennial weed. We found that the resistance cost determines the standing genetic variation for herbicide resistance. The sexual phase of the life cycle, including self-pollination and seed bank dynamics, contributes substantially to the persistence and rapid adaptation of S. halepense. While self-pollination accelerates target-site resistance evolution, seed banks considerably increase the probability of escape from control strategies and maintain genetic variation. Combining tillage and herbicide application effectively reduces weed densities and the risk of control failure without delaying resistance adaptation. We also show how mixtures of different herbicide classes are superior to rotations and mono-treatment in controlling perennial weeds and resistance evolution. Thus, by integrating experimental data and agronomic views, our theoretical study synergistically contributes to understanding and tackling the global threat to food security from resistant weeds.

Assessing the probability and time of weed control failure due to herbicide resistance evolution

“…Seed banks preserve genetic information of forgone selection for resistance and prolong the period in which a weed can adapt towards a herbicide, emphasising the complications seed banks impose on managing herbicide resistance. A previous study showed that a seed bank substantially delays the extinction of a herbicide-treated weed population and increases the risk of resistance (Lauenroth and Gokhale, 2023). Further, a seed bank buffers the genetic effects of selection, slowing resistance evolution if it occurs.…”

“…The mathematical representation of the model is shown in Figure 1 b. In Lauenroth and Gokhale (2023), we presented a deterministic model of Johnsongrass' life cycle capturing all life-history stages and intraspecific competition between the plants, allowing the prediction of long-term population dynamics ( Figure 1 a). In the present study, we are interested in the probability and timing of herbicide resistance evolution.…”

“…In the following, we assume a diploid genetic architecture with one allele conferring herbicide resistance at a single biallelic locus (Lauenroth and Gokhale, 2023). We consider the case of incomplete dominance of resistance, with homozygous resistant plants not affected by the herbicide and heterozygous plants being killed with an efficacy half as high as for sensitive homozygotes.…”

“…We suppose simple Mendelian inheritance with mutations in both directions occurring with probability 𝜇. The inheritance matrices in the sexual reproduction of plants are then given by, 𝑎 In the simulations, the limits of seed and rhizome bud numbers produced per plant are 0.93 𝑓 max and 0.93 𝑏 max to compensate for the fecundity reduction by (1 + 0.1 𝐴 −1 ∑ 𝑚 𝑖=1 𝑍 𝑛, 𝑖 ) −1 apparent even at extremely low densities (Lauenroth and Gokhale, 2023).…”

“…The parameters displayed in grey are varied in some parts and take the values given here if not stated differently. Detailed explanations and references are given in the Supplementary Information of Lauenroth and Gokhale (2023).…”