Foliage type and availability alters the movement behavior of late instar European Lymantria dispar (Lepidoptera: Erebidae)

This is the second project I worked on during my Master's work in Entomology at the University of Minnesota. By this point, I was much more comfortable with R than I was during the first part of my thesis work. At this point I'd taken three graduate level statistics classes in statisical analysis and an advanced programming in R class, which introduced me to shiny. I don't demonstrate my shiny skills here (see my MapFIA project for that). I think you can see a definite improvement in my coding ability in this project compared to my buffer zone project. My data and code are better organized. This was also the point where I figured out how to use the apply family of commands! You'll also find I started to use some of the dplyr and other tidyverse functions to summarize my data.

In these scripts you'll find code for:

• Applying the same data cleaning to a list of data frames
• Custom functions for calculating movement variables given x, y coordinates (e.g. velocity, turn angle)
• Syntax that uses dplyr to summarize data
• Statistical analyses of movement data using ANOVA and generalized linear models
• Publication quality plots produced with ggplot2

The paper can be found here: https://link.springer.com/article/10.1007/s10905-019-09711-2

Background

In an effort to better understand the movement ecology of L. dispar caterpillars, I raised them on different types of food and under different food deprivation protocols and then put them on a servosphere. A servosphere is a cool piece of equipment that's basically like a treadmill for insects. This treadmill consists of a large sphere that sits on top of three orthogonal motors. A camera positioned above the sphere monitors the insect on top and as the insect moves, the software for the sphere takes information from the camera and directs the motors to rotate, moving the ball and keeping the insect on top. The software records the position of the insect as it moves so you can reconstruct its movement path. You can export this data to a .csv file and using the position data, calculate movement variables like velocity and turn angle.

I raised the L. dispar caterpillars on four different types of foliage from trees and artifical diet that is commonly used when raising the caterpillars in the lab. The four trees I took leaves from cover the spectrum of host preference for the caterpillars. There are lot of papers that show how the host type an insect feeds on affects their performance, but not many that look at how host type affects movement. I also starved the insects for 0, 24, or 48 hours prior to putting them on the sphere to see how they responded to the starvation and if that response differed depending on what food I raised them on.

Conclusion highlights

1. Caterpillars that feed on more suitable or preferred hosts may be better equipped to search for food during periods of food deprivation. Caterpillars that fed on foliage (as opposed to the artificial diet) had the same initial capacity for movement. Only caterpillars that fed on bur oak, a preferred food source, moved farther as they went longer without food. This increase in distance moved by the bur oak-fed caterpillars was mainly due to an increase in mean velocity as starvation time increased.
2. Feeding on less preferred or less suitable hosts increases the probability caterpillars will move. Caterpillars that fed on larch and artificial diet were more likely to move than those that fed on bur oak. This response, however, is also affected by starvation time; as starvation time increased from zero hours to 48 hours, all caterpillars became likely to move.
3. Managers may need to consider adjusting the size of the buffer zone depending on the type of host plants on the perimeter and expected populations of L. dispar. Buffer zones with preferred host plants along the perimeter may have a lower likelihood of intrusion by caterpillars than those with less preferred hosts nearby. During outbreaks of L. dispar, it may be more likely that caterpillars will move into the buffer zone as they deplete food sources in the immediate vicinity.