Current camera-aided methods of assessing fish populations may be underestimating numbers by as much as 60%, Australian researchers have demonstrated.
Current practice for remote visual monitoring of marine life involves setting a bait and pointing a single camera at it.
A team of scientists led by biologist Sasha Whitmarsh from Flinders University in Adelaide wondered whether such set-ups introduced biases into the results.
To test the proposition, they set up a standard camera trap, but augmented it with several small GoPro camera units which together produced all-round vision.
“We showed that the 360-degree field of vision captured the previously unknown activity to the back and sides of the bait, giving us a better idea of abundance and diversity of species at the site,” says Whitmarsh.
“The single camera also was not able to accurately count all the fish converging on the bait because many were obscured in the crowded space. The extra cameras allowed us to count more of these individuals giving us better abundance estimates.”
The extra cameras were particularly useful for revealing the presence of “shy” fish species, which approached the bait but did not join in with the fevered competition for it demonstrated by more robust species.
In a paper published in the journal Royal Society Open Science, Whitmarsh and colleagues report that using the multi-lens rig, abundance estimates for some species were revealed to be as much as 60% higher than previously estimated.