Tuesday, April 9, 2013

Under pressure

In the United States on Februaury 2nd every year, if a groundhog sees its shadow then there will be 30 more days of winter, if not then spring follows immediately. Folklore in relation to atmospheric conditions has been common place for centuries and in times before accurate weather prediction could be made, people had a need to find ways of figuring out what was going to happen to their crop or what the conditions at sea would be like. Although most of those early ideas and myths are just fairy tales, like the groundhog story, many are steeped in truth as they were based on multiple observations over time. For example, “ no weather is ill, if the wind be still,” when there is a high pressure system, there is cooling air moving towards the earth’s surface, cool subsiding air is then warmed which may cause evaporation of clouds and often results fair weather.
Similarly, there have been stories over the centuries of animals being able to “predict” conditions such as violent storms and earth quakes. Dogs howling incessantly, caged birds becoming restless, the sound of insects disappearing as they had all left to find safety elsewhere. These are anecdotal stories but scientists are starting to research the validity of such claims. 

Here at Oceans Research it was observed on our regular chumming trips (where our interns get the opportunity to come face to face with white sharks as they help us collect essential data on population dynamics of this apex predator), that on the days after heavy storms, there seemed to be less sharks in the bay. And here at Oceans Research we are all about the research and not the anecdotes.  With a little searching we came across a paper that was published about juvenile blacktip sharks, who left their sheltered nursery grounds for deeper water before the tropical storm Gabriel made landfall, and all the sharks returned after the storm was over. It was concluded after looking at all possible environmental variables that the sharks were responding (perhaps innately) to a drop in atmospheric pressure. The only other paper published in relation to sharks and pressure showed that vestibular hair cells in dogfish are responsible for detecting changes in pressure and that these sharks can detect pressure changes as low as 5 millibars.

We have extensive data sets on the movements of white sharks in Mossel Bay, with more data coming in everyday using more and more sophisticated methods for tracking sharks such as acoustic telemetry, where sharks are tagged and are detected when moving past receivers in the bay (part of the Ocean Tracking Network- OTN), or by satellite tracking devices (part of the Ocearch expedition). The goal of this part of the project is to compare the movements of the white sharks in relation to changing atmospheric pressure systems. This is the meat of my MSc project and hopefully the results will yield some exiting information about these majestic creatures.

Another observation was made by an aquarist who worked at the Shark Lab and aquarium of Mossel Bay (more info can be found on this blog about the Shark Lab and its workings). He noticed that during storms the three species of benthic (or bottom dwelling) sharks that are on display seemed to group together more than usual when there was a storm outside. So sticking with the shark/pressure topic we decided to test this.
When it comes to scientific experimental design there can be a lot of ups and downs. What was supposed to be a simple experiment turned out to be one of the most difficult of the Master projects, with the irony being when you finally get it right you end up moving back to the simple instead of the complicated, after having discarded all the other parameters.
When looking at the grouping behaviour of sharks in a public aquarium, you have to take into account the aesthetics of whatever you are doing. You have a small sample size of sharks to work with at any time, the roof is low and the tank is wide, so photographic or video footage has to be taken with a wide angle lens. Can we correct for the fish eyes effect of that lens? 

    Figure 1: Using software we can remove the fisheye effect and measure accurate distances between         sharks.

How often are you going to record behaviours. We started at three times a week, for three hours spread over a day. But what about behaviour at night? It is presumed that some of these species are nocturnal.  Do we really need behaviour metrics such as swimming patterns when these sharks spend most of their time resting in one place? Do we somehow divide the tank into quadrants and see which are more favourable?

    Figure 2: An early experimental design

There is a skylight and the sharks tend to be on the opposite end of the tank to the skylight. Are they moving away from the light or is it because the tanks water inflow pipe happens to be on the same side as the skylight? Do we take readings on all days to encompass a more “natural” behaviour or do we avoid the three days a week that the sharks are fed and become most active? Do we need to identify individual sharks by using some sort of marking method or do we look at what they are doing as a collective bunch? The three benthic species we have in Mossel Bay (in captivity and on the reefs) tend to group and lie on top of each other as if they didn’t know they are separate species. With each answer comes a new idea or question and it compounds so much in trying to design the perfect experiment where all variables needed are accounted for and those that are not are discarded. 

    Figure 3: Redesigning

In the end Occam’s Razor often proves to be the way to go.  It basically states that among competing hypotheses, the one that makes the fewest assumptions should be selected. And when designing an experiment, a similar idea can come through, where after removing all that is unnecessary in what is needed in finding the answer to the question, we are left with a simple experiment that has all the scientific power necessary to get the research question answered. It also helps when going through the literature that you find an experiment that someone else has done which answers even more of the methodological questions you were struggling with and helps make things even simpler.
So in the end we are looking at the distances between the sharks at half hour intervals over a 24 hour period once a week to ascertain how they are grouping, and then comparing that grouping with how much they congregate during low pressure systems. We are currently in the control phase and are eagerly awaiting the stormy season to see what results we get.  

    Figure 4: Almost there

It is also  important to realize for young budding shark scientists and prospective interns which no doubt will read this blog when looking into coming to Oceans Research to learn about sharks, that their dream job isn’t just about being on a boat all day and playing with sharks. Research can be tedious and monotonous with many hours spent off the water problem solving and conducting experiments which seem extremely boring, but it all adds up to the great database of life, where scientists publish their findings and contribute to the global knowledge and ultimately the protection of the planet and the organisms which inhabit it.

Tristan Scott
MSc candidate. University of Pretoria/Oceans Research

No comments: