Fig 1.
Map of Seal Island (A) in Mossel Bay, South Africa (B), highlighting the specific location of testing sites around the island (A1-4).
Testing site locations are not exact, but, instead, mark the approximate area that trials were concurrently conducted.
Fig 2.
Diagram of a Remote Monitoring Research Apparatus (ReMoRA).
(A) Shows the ReMoRA in its deployed configuration with downward facing cameras. (B) Shows the measurements recorded to calculate proximity of C. carcharias to the visible Shark Shield™ electrode. Using Event Measure software, the closest part of a shark’s head to the electrode is marked via the left and right video clips (a), followed by the centre of the Shark Shield™ electrode (b). Event Measure compares the length and angle of the lines drawn in the left and right synchronised/calibrated video clips (c) to accurately calculate the closest observable proximity of the shark in three-dimensional space, taking into account both the vertical and horizontal axis. For clarity, the electrodes of the Shark Shield™ are displayed in white to highlight their position.
Fig 3.
Synchronised video screenshots from the left and right cameras of a ReMoRA, which show C. carcharias interacting with the bait container during a control trial.
The screenshots shown are three-dimensional, so for reference to size see Fig 2B. Using Event Measure software, the distance between a shark’s head (a) and the centre of the Shark Shield™ electrode (b) is calculated by comparing the length and angle of lines drawn between these two points on the left and right synchronised/calibrated video clips. The proximity of the shark in the screenshots displayed is 40 cm.
Fig 4.
Schematic of the equipment used to measure the voltage gradient of the Shark Shield™.
For clarity, the electrodes of the Shark Shield™ are displayed in white to highlight their position.
Table 1.
Comparison of the behavioural response of C. carcharias when encountering an inactive (control) or active Shark Shield™.
For more detailed data, see S1 Table. Justification for the statistical tests used is provided below.
Fig 5.
Schematic representation of a user wearing a Shark Shield™ with the estimated average deterrent thresholds of C. carcharias overlaid.
The dashed line depicts the average proximity of a shark’s first encounter (131 cm [9.7 V/m]); and the solid line depicts the average proximity of all encounters (82 cm [15.7 V/m]).
Table 2.
Comparison of the behavioural response of C. carcharias between sharks, and between encounters, during control trials.
Justification for the tests used is provided below.
Table 3.
Comparison of the behavioural response of C. carcharias between sharks, and between encounters, during active trials.
Justification for the tests used is provided below.
Fig 6.
Bar graphs show the proportion of sharks that interacted (grey bar) during each encounter with a control (A) or active (B) Shark Shield™ treatment.
Overlaid is the average proximity of sharks to the Shark Shield™ during each encounter (± Std. Error). Proximity trend line (Control): y = -20.876x + 323.56; Proximity trend line (Active): y = -116.37x + 1283.
Fig 7.
Plot to show the Shark Shield™ voltage gradient decline with increasing distance.
The dashed arrows indicate the average proximity/encounter (82 cm) of C. carcharias and the corresponding voltage gradient (15.7 V/m); Red dots depict actual measurements recorded using the voltage gradient probe (see Fig 4). Voltage gradient curve plotted using Harris model: y = 1/(0.0101 + 0.0003x^1.1706).