Final Summary: Modeling Milkweed Population Dynamics

Since the summer research session is coming to an end and the first major phase to my project is almost “finished”, this seems like a good time to write my final post, wrapping up what I’ve been doing all summer, what has come from my work, and how this will transition into the next phases of the project. Just as a friendly reminder, this project is focussed on community ecology and population dynamics of the Common Milkweed, with the specific goal of modeling the size and demographic behavior of the population as a function of factors like herbivory and leaf chemistry. We use field-collected data and computational/statistical models in the R programming language to determine these relationships, that could inform management policies and conservation strategies.

The beginning of the summer I spent most of my time working on data cleaning, organization, and manipulation. The end product of this work is a set of automated R scripts that can check our data for errors in the field in real-time, clean and process our data swiftly and automatically, and then join them together with past data to be used in our models. This has and will continue to save us time and effort, and has already caught more errors than by looking over the data manually, creating a higher quality standard.

Next, field season began, so we spent several weeks out in the field collecting data on milkweed plants. This includes measures of size, demography (survival/reproduction), herbivory, leaf chemistry traits, genetics, and spatial orientation. These are the essential data used in constructing our statistical models of the population. We have yet to analyze or incorporate much of this data, so this could (and hopefully will) provide answers to exciting new questions that will be part of the project further down the line.

The last part of the summer, I spent all my time coding and comparing different models to predict the vital rates (survival, growth, reproduction) of the plants in our population of interest. These models are now being incorporated into an analysis of the effect of herbivory on population growth in Common Milkweed (see here for more details).

I am reaching a good stopping point in the project after this summer, and have learned an absolute ton! However, this is not the end of the milkweed model project, and there are many other aspects planned for the near future. In the coming semesters, I will continue to work with Integral Projection Models for milkweed using the data we have collected, and start to incorporate some of our previously untouched data. I have already begun work on a new set of models that we will use to compare different techniques for measuring herbivory. The hypothesis is that the way ecologists typically measure herbivory might be obscuring its real effects. The second, and most exciting extension of this project will be incorporating leaf chemistry traits (like nitrogen, fiber, and defensive chemicals) into the model, to look at their impacts on herbivory and population dynamics. We will be using a novel approach to be getting at a question that has wide implications for the study of plant-animal interactions, restoration practices, and evolutionary ecology. Needless to say, there are lots of exciting things planned!