Green filters and a green flash

For me, the most exciting moment of a cruise is the moment the first CTD hits the surface of the water. The second, a few hours later (if you’re lucky!), is the moment the first samples hit the -80 degree Celsius freezer. This means my sampling is well and truly underway, and we’re on our way to answering some of the big questions we have about the ocean and the Earth as a whole.

This happened today at 0100 hours. We typically meet in the ship’s main lab about half an hour before the CTD is due to go in, confirm how deep we're going, what everyone needs from it, and sort out any last minute sample collection prep. Then, once we see it go over the side we head up to the bridge, where the winch operator and CTD computers are so we can watch the traces of our important parameters as the CTD descends through the water.

Left: the CTD ready to go into the water. Middle: what we observe on the computer as the CTD goes down. Right: sampling from the CTD Niskins.

Speaking of important parameters, let’s have a quick CTD 101. We use the term ‘CTD’ to refer to the rosette of Niskin bottles we use to collect seawater from up to thousands of metres deep in the ocean. But, technically, ‘CTD’ refers to the sensors attached to the frame of the rosette, which measure (amongst other things), Conductivity, Temperature and Depth. Conductivity tells us about the salinity of the water (how salty it is), temperature is self-explanatory, and we calculate depth from a pressure sensor. In addition to this core CTD, we also have a PAR sensor for Photosynthetically Available Radiation (ie light level), an oxygen sensor, a fluorescence sensor which detects chlorophyll, ie how much phytoplankton are in the water, and measures of beam attenuation, which tells us how much particulate matter (ie carbon) is in the water.

So, we watch these traces as the CTD descends, and then decide, based on these parameters, which depths we want to take water samples at. As we lower the CTD, all its bottles are cocked open, allowing water to flow freely through them. When we reach a depth we want to grab a sample from, we press a button to ‘fire’ a bottle, which snaps closed a quick release wire and encapsulates 12 litres of water at that precise depth, which we can empty at the surface for whatever analysis we please.

In my case, I’m sampling for the DNA and RNA in the water. Which brings me to the question almost everyone on this ship has asked me: why on earth am I sampling in the middle of the night?! Well, there are a couple of reasons for this. The main one is that I’m sampling communities from the deep water – up to 200 metres. As we’ve just learned, there are different light levels through the water column. Since I’m looking not just at which genes are present, but also which genes are switched on, if I were to bring those communities up into the bright daylight, they’d become extremely stressed (the same way you or I would if someone switched all the lights on in the middle of the night!) and I’d get a completely messy signal. The second reason is because plankton exhibit diel cycling – undergoing physiological and ecological changes on a 24-hour cycle, and I want to be sure I’m sampling at the same point in this cycle each time, so I don’t introduce additional variables into my analysis.

The CTD came over the water around 2 am, and thus ensued a flurry of sampling, filtering and fixing. The chief scientist and one of the BATS techs very kindly helped me to collect my water from the CTD, and then it was party for one in the aft lab until just after 4 am, whereby all samples were in the liquid nitrogen, in the -80 degree freezer, or safely tucked into the fridge.

Left: these are my filters, the end product of my filtering. Each one has had 10 litres of seawater filtered through it, and contains millions of precious phytoplankton. Right: the sunset over the North Atlantic.

My midnight sampling schedule means I head down for another few hours of rest until about 10am, at which point I'm up with the rest of the scientists for other general time series sampling, which today included the recovery of a 12-hour production array. This entry has already rambled on enough, so I'll save details of that for another day. In the evening there was another very exciting first, as you might have guessed from the title, I saw my first green flash!