Following on from my post on Saturday regarding the scientific studies done on the effects of chumming elsewhere in the world, a study done in False Bay in 2004 has been brought to my attention by the team at Shark Spotters (thank you!). Much more should be made of this work!
Effects of provisioning ecotourism activity on the behaviour of white sharks Carcharodon carcharias – LaRoche, Kock et al (Marine Ecology Progress Series, May 2007)
This paper (summary here, full text here) describes a study done in June-October 2004, using acoustic tags and receivers on the sharks and sea floor around Seal Island. Over the five month period, the researchers were at sea for an average of 15 days per month, ten hours per day. Despite this their sample was too limited to draw conclusions about certain things – this is an excellent demonstration of how hard it is to do science in an ecosystem.
The findings are similar to those of the Australian study I mentioned on Saturday in that it was observed that the chum had an effect on shark behaviour, but that it was fairly minimal. The authors estimate that 10-20 sharks are present around Seal Island at any one time during the winter (May – September), but only a small fraction of these approach to investigate the material placed in the water to attract them to the boats.
Moreover, the sharks’ response to the presence of chum decreases with time – during the first hour that a shark was recorded on one of the acoustic receivers, the chumming appeared to have a significant effect on its location at the island. In subsequent hours, however, the sharks seemed to lose interest and resume ordinary behaviour. This effect was also observed on a larger time scale, with decreasing interest over a number of days.
No effect was observed on their predation rate (how many seals they caught) suggesting that (even when they got hold of the bait being used) the chum didn’t cause the sharks to substitute their usual diet with the bait or with some other source. Sharks were seen at the boat only 36% of the time that they were actually at the island. Sharks did swim closer to the surface when the researchers were chumming in the area, but resumed their normal swimming depth elsewhere around the island.
These results are the exact opposite to what one would expect if the sharks were becoming conditioned by the presence of tourist boats chumming at Seal Island. Conditioning refers to changing behaviour by means of some sort of stimulus, in this case by rewarding the shark with a smell or taste of chum (the stimulus) each time it approaches a tourist boat, and thus causing it to approach all boats (for example – the response). If sharks were becoming conditioned by the chumming, one would expect the time they spend around the boat to increase over time, and that they might even approach boats that were not placing any attractant in the water.
There is still work to be done, however:
Unfortunately, conditioning is not the only way that chumming can directly affect the sharks. Extra provisioning could potentially alter residency times at the island, in either a positive or negative direction. It could also theoretically affect shark population structure around the island, if dominant and subordinate individuals react to the chum in different ways. However, despite the fact that it would seem reasonable to surmise that the patterns observed in our data would not translate into changes in shark residency times at the island, nor could the impacts of sparse provisioning have substantial effects on population structure, the short-term nature of the present study makes it impossible to draw any inferences regarding these topics.
I’d suggest you read the original paper if this kind of thing interests you. Even if the statistical jargon (Komolgorov-Smirnov tests, t-tests and the like!) is mysterious to you, it’ll give you a good idea of how arduous it is to conduct a study of this nature (750 hours at sea, in winter, anyone?), how frustrating it can be (the presence of cage diving operators at times meant that their observations had to be discarded in the interests of statistical rigor), and what goes into analysing the data. There is no real place for “gut feel” – it can lead your enquiries in a particular direction, but the final analysis is a statistical, evidence-based one.