One of the fun parts about field research (and much lab work too) is making equipment for really obscure purposes on a tight budget. My pond enclosures, for example, are made of window screen sewn together with fishing line. This year the DIY component of my experiment had slightly wider applications and could potentially be a fun thing to do if you have a fish tank or a favourite local watershed that you want to film.
I originally decided to film the stickleback mate choice trials so that I could figure out how long to run them for. Last year, I had no fish spawn in the first 30 minutes, but more than expected spawn after 24 hours (including way more inter-species mating than there should have been). I decided to try four-hour trials but hoped that I could eventually cut it to two hours if most spawning took place before then. Since I didn’t want to disturb the fish during the trial, filming seemed like the best option. Filming the trials would also potentially give me more data—I could record not only whether spawning took place but when, and how many times the female checked out the nest before deciding to spawn. So I needed to figure out a cheap way to waterproof a webcam and anchor it next to a nest.
My supervisor found this tutorial. It’s a pretty neat camera housing design, really cheap, and surprisingly easy to do—I used a Swiss army knife for virtually every step. I bought the cheapest webcams I could find, and used some old body lotion containers. I cut a window in each of these and covered it with a piece of an old CD case. The trickiest part was rewiring the webcam – I had to make a hole in the housing for the cable to go through, then cut the cable, thread it through the hole, and reattach all the wires. To keep the lens from fogging up, I put some silica gel inside the housing to absorb moisture. I glued everything up with aquarium silicone (which woudn’t leach harmful chemicals into the pond). Since the housing was full of air, it was positively buoyant, so I attached them to dive weights to keep them on the bottom of the pond. The cameras were connected via a bunch of boosted USB cables (actually the most expensive part of the project!) to a laptop on shore.
The results were mixed. The cameras worked as well as could be expected (they were, after all, dirt cheap). But positioning the camera so that it faced the nest was extremely difficult. I tied a piece of fishing line to the housing so that I could move it around without pulling on the cable (which would dislodge the silicone and cause a leak), but this often ended up slipping off the camera—so I had to pull it up by the cable anyways. Fortunately, the first few times they leaked I was able to dry everything out and reuse them. Eventually, however, the cameras just died. Possibly next time I should try drying them out in a jar of rice instead of just air-drying. (“Next time”, yeah right.) I also probably could have come up with a better way to attach the weight and fishing line, but I decided not to worry about it. The positioning problem made it clear that I wasn’t going to get much data from the videos, so I changed my goal to just trying to get some decent video I could show in a presentation. (Research is all about redefining goals, a.k.a. lowering your expectations.)
I did get some statisfactory footage. Here’s a benthic male nesting in the open (ish). I ended up not using this trial because the female wasn’t actually ready to spawn, and, as you can see, the male’s nest is actually in sparse vegetation that I had overlooked when I put him in the enclosure. It shows some of the nesting behaviour well, though. The male is in bright nuptial colouration—blue body and iris with a bright red throat—and he pokes around at the nest and occasionally deposits spiggin. At about the 1:50 mark, he also swims through the nest, which is pretty adorable. (Oh, and there’s a backswimmer kicking around in there too.)