Florida Turtles and the Mayan Collapse

By Laura Brenskelle

Here is a riddle for you: what do turtles found at an archaeological site in south Florida have to do with the so-called “collapse” of the Maya in the Yucatán, Guatemala, and Belize?

… Do you know the answer yet? I’ll give you a hint.

Well, nothing really. So, why did I spend the end of last semester downloading paleoclimate data, counting turtle records, and figuring out how to break up the chronology of a single mound at the Fort Center archaeological site in Florida if it’s completely unrelated to the Mayan collapse project that I am working on for my fellowship? Because of the process.

Image from http://www.userexperience.co.nz/

At the time, I was finishing a graduate statistics course and needed to complete a semester project, but I did not yet have data in hand for the specimen records at the Mayan sites. What’s a graduate student to do? In my case, the answer was to find a site with accessible data, scale down the project so it was doable over the course of about two weeks, and use this as an opportunity to familiarize myself with the process of doing this kind of modeling, statistical analyses, and research. I have the ability to do this because I am a biodiversity informatician. I have never done archaeological fieldwork in the Maya region or Florida or anywhere at all; my expertise is in the reuse of data that others have collected. With enough contextual information from archaeologist collaborators, I was able to figuratively jump from the Maya region to the Fort Center site in Florida. The purpose of this was not only to fulfill my statistics course requirement and learn something about the relationship between turtle occurrences and paleoclimate at Fort Center, but it actually served as a mini-example of how I can tackle the Mayan collapse project now that I have specimen data.

 What did I find? Unfortunately, not much. The temperature and precipitation paleoclimate records at Fort Center did not vary much during

Figure 1: This graph shows the turtle counts (black circles), the average annual temperature (red line), and average annual precipitation (blue line) for each time range. As you can see, there is not much variation in the climatic conditions in this region at this scale.

the time frame that I had turtle specimens from (Figure 1). However, despite the lack of significant results, the process of doing this project was enlightening because it helps me not only to better understand the process of running these analyses, but also the kind of data I will need in order for these analyses to be informative. Full disclosure: the Fort Center data were not perfect, but we almost never get perfect data. I had to estimate time ranges for when the turtles came from over the course of Fort Center’s occupation. This meant the time scale for my project was quite broad from 200 to 1700 AD. This in turn meant that I had to basically use average annual temperatures averaged again over large time spans (250 years to be exact). Given all of this, it is not surprising that the Fort Center results were not significant. However, it is worth noting that this time period of occupation at the Fort Center site is thought to have been climatically stable. This is not thought to be the case for the Maya region during the proposed “collapse”; hopefully this means the paleoclimate data in this region have more variation.

 Unfortunately, the status quo in science is to only share flashy, significant results. If that’s so, why am I telling you about a “failed” project where no significant results were found? Despite what the public may believe and what scientists may hope for, studies like my Fort Center mini-side project are far more common than those with flashy and significant results. They actually serve a very important role in the scientific method by allowing us to analyze why these “failures” failed. They allow scientists like myself to test new methods that they are unfamiliar with. Over the course of this semester, I will take what I learned from this Fort Center work and apply it to the Maya region. I will also be scaling up what I did, and focusing on multiple types of animals with many sites to look at variation across the region.

 A lack of significant results does not always mean that a project was a waste of time. Science is not always a linear process. Sometimes small side steps or taking the time to jump to a different task can actually help move forward another study when you are having difficulties.