My current research, for laypeople

I have struggled a lot to explain my current line of research to people in my life who aren’t scientists. But if I can’t explain my research to everyone, then I can’t claim to really know what I’m doing. So I’m going to try my best.

Models of natural phenomena – say, the size of the elephant population in a nature reserve – usually have some kind of randomness in them, because natural systems constantly change. Randomness is usually modeled as white noise, like this:

white noise

The problem is that randomness in natural phenomena is usually not white noise, but colored noise. See, the randomness of white noise has no “memory.” Each point is uncorrelated to the point that came before it. But that’s not how it works in the real world. If the water level in a lake is high this year, it’s more likely to be high next year. The water level is still fundamentally random, but conditions tend to carry over from year to year. This kind of randomness is called red noise.

red noise

The other type of colored noise is blue noise. With blue noise, each point tends to go in the opposite direction of the one before it. This is like masting in trees, where they tend to make a lot of pollen and a lot of flowers one year and much less the next.

blue noise

When you look at randomness in natural phenomena, like the number of elephants in a nature reserve or the water level of a lake or sea surface temperatures in the North Atlantic, they tend to have red noise. If you want to make a model of a natural system that includes randomness, it’s more realistic if you model that randomness as red noise. Now, a realistic model is not always the best model. Sometimes a simple model is better if it does a good job describing the natural system using less information. But I wanted to know whether a particular type of population model – a model of how an animal or plant population grows and declines over time – is improved by using colored noise instead of white noise.

To do that, I made a open source software package in the programming language R called colorednoise. This package can create colored noise and run population models with colored noise. I’ve been using the package on plant and animal population data from the COMPADRE and COMADRE databases, which make these data publicly available for scientists to use in ecological research. I’ve finished the study already, it’s just a matter of writing it up into a scientific article at this point!

I'm going to Denmark! Here's why.

Julia with diploma

With my Master's degree in hand, I'm happy to say that I will be starting a year-long fellowship with the European Doctoral School of Demography (EDSD) in September, at the Max Planck Center for Biodemography in Odense.

When I tell people about this exciting new chapter in my academic career, they usually ask me, "What's biodemography, and why do you want to study it for a year?"

Demography is the mathematical study of populations. Populations can grow and change in a variety of ways. Individuals can immigrate and emigrate. Some are born and some die. The population is composed mostly of old trees or mostly of saplings.

Demography is used a lot by social scientists to study trends in human societies, but when you use demography to answer questions about biology, that's biodemography. Demography can give you insight into a lot of different things: how populations respond to environmental disruption, how patterns of growth and reproduction vary from species to species, the evolution of aging, and how to save an endangered species or stop an invasive one.

My fellowship is going to be computational and quantitative. I will learn mathematical techniques to analyze population dynamics and apply them to demographic data from species that interest me. I'm not sure which species yet, but they almost certainly won't be human. I'm interested in using demography for conservation, and humans aren't exactly an endangered species.

How my research applies to humans (if at all)

One of the more common questions I get from people when I tell them that I study cooperative behavior in paper wasps is whether my research can teach us something about cooperative behavior in humans. My first thought is always that I, personally, don’t really care whether it applies to humans or not. That’s not the reason I got into this field. At this point, we don’t even fully understand why social amoebas cooperate with one another, so to me it seems that the goal of understanding why humans cooperate in the ways we do is a very long way off.

But I usually don’t say that first thought out loud, because I don’t want to dismiss people’s desire to turn to other animals to learn more about our own species. That desire drives a lot of good scientists to do good research, and gets a lot of lay people interested in animal behavior. I want to encourage that.

I think it’s easiest to explain how my research contributes to understanding human society by employing a simile to another field of biology: cancer research. When people donate money to cancer research, they often imagine that the money goes directly to developing new cancer drugs. And sometimes it does. But a lot of cancer research would look to a lay person like it doesn’t have anything to do with cancer at all. Lots of cancer research is done on DNA replication and repair proteins in yeast. Yeasts don’t even /get/ cancer. But this research is still crucial to understanding how cancer works in humans, because cancer is fundamentally cell replication gone out of control. It happens when replication mechanisms in the cell go awry, and those mechanisms go awry when mistakes in the genetic code don’t get properly repaired. The genes in yeast that control DNA replication and repair are actually quite similar to ours – which is why biologists do cancer research on organisms that don’t get cancer.

The social life of a paper wasp is very different from yours or mine. But there are very simple mechanisms that influence their social lives that also influence ours, such as aggression, discrimination, and nepotism toward relatives. I’m pretty sure that if we don’t understand how these mechanisms work in wasps, whose societies are much simpler than ours, we have no hope of understanding them in people.

My 2016 fieldwork adventure begins

I’m driving to Virginia on Friday, where I’ll be staying at the University of Virginia’s Blandy Experimental Farm for three weeks. Here’s a map of all the sites where I hope to search for wasps during my stay.

map of 2016 VA & MD fieldsites

Starting on Sunday, I will wake up around 4 a.m. every morning to go out and search for wasp nests at farms, stables, and parks. Thanks to the good folks at Virginia Working Landscapes, I already have most of my first week scheduled. Tune in to my Twitter for day-to-day updates on the status of my research.

Surveying wasp nests - how I'll do it

I’ve been taking some time to think deeply about how to collect the data I’ll need for my dissertation project about geographic variation in cooperative breeding in Northern paper wasps, and I thought I’d share with you all my current plan.

When I pick out a field site, I’ll go there four times during the spring and summer. The first time will be in late April / early May. I’ll need to visit the site twice. The first time is to survey for nests, which I can do at any time of day. The only equipment I’ll bring is a pair of binoculars, a laser rangefinder, and a ladder to get a better look at nests if I need to. The amount of time this takes depends on the size of the property; it takes ten minutes to survey a small barn, took me eight hours to survey all of the buildings in Acadia National Park.

Assuming I find nests at the site, I will then do a second visit, which needs to happen in the early morning, ideally 4 am but I can do it later in the morning, as long as it’s before 8 am. Some of the nests I will collect and freeze, and some I will keep in situ and mark the wasps with paint so I can identify them later. The equipment I bring with me for that is a ladder, a wheeled cooler, paint, and forceps for grabbing the wasps with.

I need to do three follow-up visits in June, July, and August. For those I only need one visit because I know where the nests are and I don’t need to survey for them, but the visit does have to be early morning. For the June and July visits all I’ll do is visit and observe, no equipment. For the August visit I’ll collect again, with my ladder and cooler, and round up all the reproductive females produced on the nest this year.