PhD Research

PhD thesis: Distributions and interactions of insect herbivores as influences on host plant density and performance

I completed my PhD in July 2012. I was in the Department of Zoology and Biodiversity Research Centre at the University of British Columbia and supervised by Judy Myers and Diane Srivastava

Insects damage can reduce plant performance. Reductions in plant performance can reduce plant fitness and may lead to plant population decline. My research asks what factors increase the impact that insects have on their host-plants. These factors might be traits of either the insect or the plant, or might be an external factor (either abiotic or another member of the community).

This question is clearly of interest to weed biological control and I used a weed biocontrol system (that of diffuse knapweed, Centaurea diffusa) in some of my PhD chapters.

I asked a series of questions about factors which might alter the impact an insect has on a plant – details below. A side project looked at how the plant community changed following successful biological control of diffuse knapweed.

  • Resource concentration by insects & implications for plant populations

Plants occur in different size patches and insects distribute themselves across those patches in a variety of ways. The Resource Concentration Hypothesis proposes that herbivores are more abundant per unit plant at higher host plant densities.

Using a simulation model, we show that differing distributions of insects that can exert top-down control on the plant population, can alter density dependance of plant popuation growth and the extent to which plant populations persist.

More information

Link to paper

  • Influences of two life-stages of the weevil, Larinus minutus, on its host-plant Centaurea diffusa.

In weed biological control, it is preferable to achieve control with the fewest number of introductions.

I hypothesised that using one species with two feeding modes would potentially increase the damage to plants without the costs and risks inherent in introducing a second agent.

Experimental analysis of the impacts of feeding damage by two life history stages of a successful biological control agent, Larinus minutus, demonstrates a lack of negative interactions between adult feeding on leaves and stems and larval feeding on developing seeds.

The life-history stage that reduced plant fitness differed at two sites indicating that the impact of the two damage modes covered a broader environmental spectrum than one alone.

More information

  • Interacting natural enemies: impact on the host-plant

Every plant is host to a variety of other organisms, many of which are natural enemies of the plant. Natural enemies that share a host plant may interact, even when they are separated in space and/or time.

With a meta-analysis I tested if the presence of other natural enemy alters the relationship that an enemy has with the host-plant. I asked whether the reduction in plant performance caused by two natural enemies (either plant pathogens or insects) could be predicted from their individual effects – i.e. was reduction in plant performance independent?

I showed that the reduction in plant performance caused by one natural enemy is, on average, independent of the second. I also show that the occurrence of non-independent interactions between natural enemies can be predicted by a small range of enemy or plant attributes.

More information

Link to paper

  • Testing biological control agent compatibility: Cyphocleonus achates and Larinus minutus on diffuse knapweed

While weed biological control success is typically achieved with one agent, multiple agents are invariably introduced.For biological control agents to be compatible, the reductions in plant performance should be independent and there should be no competitive interactions between the two species.

My experiment using the root herbivore, Cyphocleonus achates and the aboveground herbivore L. minutus showed that plant performance reductions of the two species are independent on each other and that there is no evidence of competition between the two species, suggesting that the two species are compatible as biological control agents.

More information

Link to paper

Side project:

The goal in biological control in natural areas is to reduce populations of the invasive weed so that native communities can re-establish. I analysed a five-year data-set which showed that after diffuse knapweed declined due to attack by Larinus minutus, it was replaced by another invasive cheat-grass Bromus tectorum, rather than native species.

More information

Link to paper