The source of green lipped mussel's spat from Ninety Mile Beach

Victoria University of Wellington - New Zealand

Aquaculture of the green-lipped mussel (Perna canaliculus, Gmelin, 1791) has recently become New Zealand’s largest aquaculture export earner. However, the industry is reliant on wild caught larvae (“spat”) that wash ashore attached to foliose seaweeds and hydroids at Ninety Mile Beach within the quota management area of GLM 9. Whilst a small industry is based on the collection and relaying of these spat to farming operations, the industry is vulnerable to reductions in supply arising from variable spat fall at this one location. Moreover, the location(s) of broodstock populations that supply these larvae are currently unknown, thus risks to the adult beds supplying spat could be better managed if the provenance of spat was known.

There are a number of methods available to determine the provenance of marine invertebrate larvae,with the two most popular being genetic and geochemical population markers. Previous work on genetic population markers of P. canaliculus has been able to discriminate North Island populations from those in the lower South Island, but has typically struggled to resolve the local population structure necessary for management. However, there is some promise in the recent development of microsatellite markers for this species which may offer the required spatial resolution to identify discrete populations within GLM 9. Alternatively, geochemical markers comprise site-specific signatures of chemical elements laid down in the shell. While this method has been shown to resolve P. canaliculus populations on the scale of tens of kilometres, this was only for populations within the Auckland region, and thus geochemical markers remain largely untested around Ninety Mile Beach.

Accordingly, our objective was to test the resolution of microsatellite and geochemical markers in discriminating adult populations along the west coast of the North Island of New Zealand. Sites selected were separated on scales of hundreds to tens of kilometres and fell within the boundaries of GLM 9 (Scott Point, Ahipara, Tanutanu Beach, Mitimiti, Whatipu) and GLM 8 (Oakura). At each site, 50 juvenile mussels were collected and the flesh dissected out for genetic analyses and the shell used to test the efficacy of geochemical markers.

We found that use of 10 microsatellite loci developed for P. canaliculus further confirmed a northsouth split in northern and southern green-lipped mussel populations. Furthermore, success of assigning samples back to their site of collection varied between 30 and 81 %, with a mean success rate of 50%. In an expanded analysis including a further 14 sites, mussels from all northern sites collected clustered together (with the exception of Mitimiti), suggestive of a high degree of genetic differentiation of northern from southern populations, and providing evidence of genetic differentiation within the northern group.

Quantifying geochemical markers within the shells of mussels revealed that eight elemental ratios (Sr, Zn, Ba, Ni, Mn, Co, Cu, and Li:Ca) could be used to successfully assign mussels back to their site of collection with between 88 (Oakura) and 100 % (Whatipu) success. Most promisingly, mussels collected from two sites separated by 4 km of coast (i.e., Ahipara and Mokorau Stream) could be discriminated with 94% success indicating the potential of geochemical markers for discriminating P.canaliculus populations within GLM 9.

However, greater power could be achieved by a larger long-term study that encompasses more sites and makes use of the different pieces of information that both these methods provide. An assessment of temporal stability in provenance of spat at Ninety Mile Beach is needed given the highly dynamic nature of the environment and breeding biology of the species. Furthermore, the use of LA-ICP-MS methods that can sample different parts of an intact shell would also be of benefit, as temporal variation in elemental signatures over the lifetime of the mussel could be quantified.

For more information see report below: