Saturday, May 23, 2015

Speak to me

So how did you spend your Friday? Mine was quite eventful. It started with an e-mail; then I crunched numbers, killed a tree, stared at the wall, and took a giant step forward.  

Me retrieving specimens from ROV Kraken II aboard
NOAA ship Nancy Foster in 2012. Photo by Megan Chesser.

You see, I've been working on analyzing data from shipwrecks off the east coast of the U.S. I got to visit the shipwrecks in 2012 with other scientists as part of a large, multi-disciplinary project. The wrecks are all located at the continental shelf break, about 100 m or so below the surface, so we used an ROV (remotely operated vehicle) deployed from a research vessel to view the shipwrecks and collect samples. I've had the videos in my possession for about 3 years now, but I'm just getting around to analyzing the data. (Yes, that kind of a time frame is perfectly normal.) 

The cool thing about the shipwreck videos is that several different researchers are working on them for different reasons. There's an archaeologist in Rhode Island working to identify each of the wrecks and re-construct the history of the ships, plus a team in North Carolina that has analyzed the fish communities at each wreck. My job is to understand the invertebrate communities, a task that has so far been both tedious and rewarding. 

I started by watching all of the videos and taking frame grabs whenever I got a good view of the wreck surface. Then I sorted the frame grabs, sub-selected the best ones, and counted the animals visible in each photo. I estimate that by now, I've spent 87 hours watching videos and 50 hours counting animals, not to mention the time I spent sorting frame grabs or taxonomically identifying voucher specimens. 

On Friday, I finally had all my data together, and it was time to figure out what they mean. I sorted all my frame grabs by whether they came from the top, the middle or the bottom of the wreck. Then I re-arranged them by horizontal, vertical, slanted, or irregular surfaces. Eight shipwrecks times five diversity indices times two ways of sorting plus splitting the species into sessile suspension feeders and mobile predators equals, well, in the end, 144 graphs. Looking at the graphs one by one on my tiny laptop screen just wasn't cutting it, so I decided to go old-school and print them all out. Then I cut them apart. Then I taped them to the wall. Then I color-coded the patterns; red for increasing, blue for decreasing, black for no significant change. I started at those graphs for way too long, silently willing them to speak to me. 

The patterns are complicated and vary by wreck, but in the end, I think I may have made a break-through. Some of the wrecks have similar diversity patterns, and when I looked back at my map, I realized they were all oriented the same direction. I don't want to give away the details because I still have to confirm what I saw, but I may be on to something. All it took was some wall space, some paper, and some imagination. The data, they speak. 

No comments:

Post a Comment