DISCo night
by Jean-Olivier Irisson, Claire Paris, on
Through this PUF-sponsored project we have been able to study the orientation of fish larvae during the day and relate it to physical forcings such the position of the sun in the sky. It is already pretty crazy to think that those cm-sized organisms are able to detect the position of the sun from underwater and orient with respect to it. However, fish larvae are known to settle on the coast at night, not during the day. So while the sun would be a useful cue for large scale navigation in the ocean, they have to use some other cue to settle at night.
But studying behaviour at night is, of course, more challenging than during the day. Only a handful studies have tackled the question and always used traps or lab experimental settings, never in situ observations. We set out to change this and observe larvae in the DISC at night. To do so, we needed a way to see without being seen. Because light wavelengths in the red color range are quickly absorbed in water, most sea creatures do not "see" the reds. We went even beyond red and used infra-red LEDs to light up the DISC observation area, just to be sure we would not disturb the organisms. But infra-red is basically heat and the LEDs heat up very much during prolonged use, which makes them difficult to attach to a transparent plastic instrument. Furthermore, while CCD sensors on digital cameras record infra-red just fine, these wavelengths are problematic for picture quality in normal-life applications and all lenses include an infra-red filter to block them out. So it took over a year of development with Bellamare to get to a working system, comprised of very bright infra-red LEDs (to shine through red-absorbing water) potted in custom aluminium and epoxy housings, a large Lithium battery pack to power them for a few hours, a modified GoPro camera with a customised lens, etc. The result is below: a nice, focused image, captured in complete darkness!
Thanks to the new extension of the project (as explained in the previous post), we could plan the first in situ observations of fish larvae at night, offshore Miami. The "DISCo night" team was lead by Professor Claire Paris and comprised Captain Evan D'Alessandro, Craig Raffenberg and Alessandro Cresci (Pr Paris' students), and Pr Jean-Olivier Irisson, from UPMC. Between July 29th and August 5th, we caught many larvae of the bicolor Damselfish (Stegastes partitus) and deployed the DISC for four nights to observe their behaviour. Instead of the usual time-lapse pictures, we captured video to get a better understanding of their behaviour. In turn, this required some non-negligible changes to discr, the software written to analyse DISC data. A few bad weather days in the middle of the week allowed to prototype those changes.
The bad weather days and the abundance of larvae were also an incentive to test the new lab-based experimental setup of Pr Paris, at RSMAS. Her wet lab is equipped with a cylindrical aquarium large enough to fit the DISC (and a PhD student, if needed), light-insulating curtains, and an array of lighting systems allowing to simulate full daylight, select some wavelengths, or use wavelengths invisible to humans such as UV or infra-red. Damselfish larvae were observed in the DISC, while it was rotating in this setup, during the day (using simulated daylight) and at night (using infra-red). The goal of these first experiments is to determine if larvae still orient indoors, which would suggest that this behaviour is somewhat ingrained and not conditioned by external stimuli. Now we need to analyse the data!
