Development of aquaculture technology for the Hawaiian opihi cellana spp.

Abstract

The purpose of this work was to develop effective technologies for commercial aquaculture of the Hawaiian limpet Cellana spp., known as opihi in Hawaiʻi. In this dissertation, our aim was to complete the life cycle of the yellow foot opihi Cellana sandwicensis. The first objective was to learn how to collect and develop an appropriate handling technique to maintain wild opihi to be used in feeding trials and also for breeding stock. The first collection trial showed a 58 % mortality, presumably due to injuries to animals while removing them from the rock. There was additional mortality of 42 % during transfer from one tank to another because they clung tightly onto the tank's wall. We were eventually able to achieve a minimum collection mortality of 30 % by being careful, however, it may be overestimated due to varying collection conditions. A holding system using plastic liners and soft plastic container led to zero mortalities due to handling in the subsequent trials. The next objective was to develop a commercial feed available for opihi. The analysis of the stomach contents found that benthic diatoms were the preferred feed in the wild and experimental feed trials on benthic ditoms called "biofilm" grown in the laboratory found that the opihi fed well at 0.47% dry matter/body weight/day (DM/BW/day) and survived well on it. Biofilm was the key to feed consumption when it was incorporated into artificial feed. While biofilm worked well as a maintenance diet, it is a natural feed that needed to be cultured and was unreliable over the long term. We found that a commercial preparation of the algae Porphyra commonly known as Nori was as attractive as biofilm and was an effective replacement for biofilm as a feeding stimulant. A formulated diet of fishmeal, soybean meal and krill meal including Porphyra produced the best growth rates (0.44±0.37 %/day) and feed intake (0.73±0.53 % DM/BW/day) over 10 weeks. Addition of krill meal combined with Porphyra into the diet produced a higher feeding rate (about 5 times) and promoted a significantly higher growth rate than those diets without krill meal. Among the experimental dietary protein levels from 21 % to 50 % and carbohydrate levels ranging from 18 % to 37 % diets were tested. Weight gain and specific growth rates of opihi increased with increasing dietary protein from 21 % diet and reached to maximum at the 35 % diet, and significantly (P < 0.05) decreased at the 50 %. The fastest growth rates of animals were obtained with 27 % (0.27 % day-1) and 32 % (0.26 % day-1) and werer significantly higher than 18 % and 37 % carbohydrate diets. We conclude that about 35 % protein and 32 % carbohydrate levels could be used for opihi and could be a model for aquatic herbivores. Monthly measurement of gonadosomatic index (GSI) and histology analysis of the gonad of the animals found that the peak spawning season for opihi occurs from November to December and seems to extend to January. The non-reproductive season is proposed to take place from late February to early September, followed by a final maturation phase which occurs in October. This finding provided an important exogenous key factor for developing an affective induction of opihi gonad maturation in the laboratory. Final maturation was successfully induced by adding ARA (arachidonic acid) into the diet at an ARA/EPA (ecosapentaenoic acid) ratio of 0.70. The GSI values of opihi reached to final maturation stage of 24.5 % and 23.7 % for 0.24 % ARA and 0.39 % ARA diets respectively, and were significantly higher than GSI of opihi (6.11 %) that were fed with the control diet without supplementation of ARA after 75 days. Poor reproductive performance was obtained when the animals were exposed to the wrong natural photoperiod. Another trial using salmon gonadotropin releasing hormone analogue (sGnRHa) at dose of 250 ng/g BW, at 7 day intervals during the reproductive photoperiod produced final maturation after 3 injections. Spawning induction by using sGnRHa at dose of 1,000 ng/g BW was an effective technique than hydrogen peroxide (0.6 x 10-2%) and was considered the most practical method because no mortality occurred after spawning. Embryonic and larval development of C. sandwicensis occurred rapidly. The larvae hatched in 12-14 hr after fertilization and subsequently achieved metamorphic competence at two days old. Larval rearing was attempted to test the survival and possible settlement and metamorphosis on different diatoms and pelagic algae but failed at Day 9. We speculate that larval mortality after Day 9 could be caused by unsuitable benthic diatoms. This led to the incomplete life cycle of the opihi in the laboratory, due to mortality of the settled larvae. More work is needed to determine if our maturation methods are truly effective, on spawning techniques to improve seed production and larval rearing in order to complete the life cycle of opihi to make sustainable aquaculture of opihi possible in the future.