Optimising the delivery of the key dietary diatom Chaetoceros calcitrans to intensively cultured Greenshell™ mussel larvae, Perna canaliculus

The diatom Chaetoceros calcitrans forma pumilum is an important dietary component for cultured Greenshell™ mussel larvae, Perna canaliculus. However concerns surrounding potentially deleterious side effects have motivated a series of experimental trials to optimise the safe usage of this alga. Bioassays involved raising veliger larvae in the purpose-built Cawthron Ultra-Density Larval rearing (CUDL) system; an array of 2.5-L tanks were stocked with 2-day-old veligers (200 larvae mL−1) supplied with inflowing water dosed with sufficient microalgae to maintain a fixed concentration of cells, after compensating for ingestion. The nutritional role of C. ‘calcitrans’ was examined by adjusting its cellular fraction in the feed environment. Diets of 0, 5, 66, 95 and 100% C. ‘calcitrans’ were each offered to six replicate rearing tanks, using Isochrysis aff. galbana (T-Iso clone) to maintain a total of 40 cells μL−1 in the larval cultures. The 66% C. ‘calcitrans’ diet was also offered at 3, 20, 60 and 120 cells µL−1 (n = 6). Higher C. ‘calcitrans’ fractions sustained faster growth, with 95% and 100% treatments producing 23-day-old pediveligers of 240 ± 15 μm and 228 ± 5 μm mean shell length, respectively; compared to 212 ± 7 μm in the 66% treatment (40 cells μL−1). However, high C. ‘calcitrans’ treatments were more volatile, had higher mortality and greater predisposition towards population crashes. Similarly, reduced survival was observed in the high feed treatment of 120 cells µL−1, reflected in a final pediveliger yield of 29 ± 6%, compared to 47 ± 3% at 60 cells µL−1 and 36 ± 7% at 40 cells µL−1. The 5% C. ‘calcitrans’ and 20 cells µL−1 treatments showed signs of nutrient limitation, while larvae fed 0% C. ‘calcitrans’ or a total of 3 cells µL−1 starved, failing to reach metamorphosis. A standard diet consisting of 66% C. ‘calcitrans’ and 34% I. aff. galbana maintained at 40 cells μL−1 was used to test the hypothesis that culture age increased deleterious side effects associated with feeding C. ‘calcitrans’. The final stage of batch culture, using 20-L nylon carboys inoculated with 1010C. ‘calcitrans’ cells was allowed to age for 2, 3, 4, 5 or 6 days before being fed to larvae. After 21 days eating 2-day-old C. ‘calcitrans’ larvae reached a mean shell length of 236 μm, were eating 35,000 cells larva−1 day−1, and had a survival rate of 59 ± 3%; in contrast, larvae eating 6-day-old cells only reached 214 µm, eating 17,000 cells larva−1 day−1, with an overall survival of 46 ± 7%. While C. ‘calcitrans’ is valuable in the larval culture of P. canaliculus and many other bivalves, appropriate feeding protocols are needed when using this diatom.