Detection in the field:
In general, where the four species of jellyfish were known to occur, their eDNA was detected. C. sivickisi was detected in 100% of replicates at sites 1, 3 and 4. At some sites no lights were used to attract jellyfish (‘blind’ samples), but eDNA from these sites was still detected (sites 10, 11). There was no correlation between abundance ofC. sivickisi medusa in the field and quantity of DNA (Figure 2, r = - 0.04, df = 23; P > 0.05).
C. xaymacana and C. fleckeri were relatively rare at the time of sampling and C. fleckeri was not detected visually, although they were known to be in the area due to observations and collection in tows from Surf Lifesaving Queensland. Carybdea xaymacana was detected by both JCAMS, as well as with snorkelling surveys; however, only at two sites and with low species abundance. That being said, this species was also still detected using an eDNA approach.Carukia barnesi was never visually detected during sampling and this concurred with zero detection with eDNA (Table 2).
A reverse thermocline was detected at 2 m with no significant halocline being detected.
When the adult medusae of C. sivickisi were not present (i.e. in winter), an eDNA signature of this species was detected in water samples. Positive detection was only made in water samples collected within 0.5 m of the substratum. The eDNA of C. sivickisi was not detected at the surface (Figure 3). Given medusae of this species are not present at the time of sampling, the eDNA signature can only be explained by the presence of polyps hidden within the coral substratum. It is likely that a shallow thermocline constrained the eDNA signature below the surface (Figure 3a).