Hairy beast

Some of the common coral genera in our study

"Every project starts out a hairy beast."

- I wrote this in a text to a friend today 

It is a sunny Friday in Woods Hole, and I am feeling victorious. Recently, I've been working on a paper about recruitment dynamics of corals in Palau. You know - the project that had me so obsessed with finding juvenile corals for 5 years. I had such a hard time getting baby corals to settle on tiles that I gave up and started looking for them on the reef. Turns out my camera was a much more effective tool for finding small juveniles than the tiles I had tried. Over the course of 3 years, my team collected thousands of images. Every single one of them had to be identified to at least genus. The project was one giant, hairy beast. 

When I finally took a look at the final data, it was messy. There were significant differences all over that dataset - between adults and juveniles, between juveniles in the photos and recruits on the tiles, between inner and outer sites, between sites in the north, middle, and south of our study area. Normally, scientists beg their data to show something, anything that's statistically significant. I had the opposite problem - everything was different from everything else. How in the world am I supposed to interpret those results? 

My first draft of the manuscript was just as messy as the data. I had no idea what I could even conclude from our study. I sent it around to the team asking for feedback - and I'm so glad I did.

Matthew left a comment in the margin of my draft, flagging one result in particular. "Hm, I interpret this completely differently" he had written. Cas had actually responded with his own comment - and yet another alternate interpretation. For starters, it's probably worth highlighting that three well-informed, competent, professional scientists looked at the same result and came to three different conclusions. More important, though, is what those comments led to. 

It forced me to nail down a framework. I made a chart in my notebook: if A, then X; if B, then Y. It was what the project needed all along. 

We hypothesized that corals would experience high mortality shortly after settling down to the seafloor - after all, most invertebrates do. That mortality could shape the composition of the coral community at different reefs. Basically, things settle on the seafloor, and as the corals that can't live in the place they settled die off, the communities come to look different. If our hypothesis is true, then we would see a high diversity of coral recruits (the ones that have just settled on tiles) and no differences between sites. As they grow, some die off, so we expect lower diversity of coral juveniles (the slightly larger ones we photographed) and maybe some differences between sites. Then we expect the lowest diversity for adults and strong differences between sites. 

Did our hypothesis turn out to be true? Well, I don't want to spoil the paper, so stay tuned. In the meantime, I'm incredibly grateful for my team of collaborators. My hairy beast of a project has turned into a docile creature. Thank goodness.