COS 109-2 - High seedling recruitment may influence long-term biological control efficacy for Euphorbia virgata (leafy spurge)

Abstract: Effective weed management with biological control (biocontrol) requires that specialized herbivores persistently target key demographic stages limiting host weed populations. However, few post-release data are available to evaluate whether long-term pressure from an introduced herbivore changes demographic structures of weed populations over time, and thus, influence agents’ control. For instance, when clonal weeds also reproduce by seed, increased herbivore pressure on one mode of reproduction may increase the importance of the other to weed population growth or persistence. Variation in the prevalence of sexual versus vegetative reproduction can therefore affect biocontrol efficacy across space or time. We quantified weed genetic diversity and density, and associated root-feeding biocontrol agent abundance (Aphthona spp. complex), in 100 populations of the clonal weed Euphorbia virgata (leafy spurge) across North Dakota, Montana, and Idaho, U.S.A. Genetic diversity provides insight into the frequency of seedling-based plant recruitment. Root-feeding Aphthona spp. agents should reduce existing stem densities and clonal recruitment, creating opportunities for seedling establishment. Quantifying diversity across many populations allowed us to examine how invasion intensity (=weed density) and biocontrol pressure (=agent abundance) was related to the relative prevalence of clonal recruitment, an assumed demographic target of Aphthona spp. agents. We found surprisingly high genetic diversity across E. virgata populations, suggesting seedling recruitment occurs regularly within a broad range of sites and augments vegetative spread to determine the density and spatial extent of local populations. Few sampled populations had low genetic diversity, though sites with lowest diversity also had low E. virgata stem densities, supporting the idea that seed-based recruitment is directly influencing weed infestation level. Associations with agent abundance were nuanced, and likely complicated by interactions with biocontrol agent response to different resource (weed) densities. Low diversity, low-stem density sites were associated with relatively low Aphthona spp. abundance, but low agent abundance was also found in the sites with the highest stem densities. Such trends require more extensive and empirical study to develop a mechanistic understanding of how agent-weed interactions impact weed demography and feed back to influence biological control efficacy. Increased population-level importance of seed production may be changing the priority of demographic targets for successful biocontrol. Further evaluation is needed to ensure the biocontrol program for E. virgata remains sufficient for long-term sustainable control.